example_id stringlengths 16 123 | messages listlengths 14 3.37k | metadata dict | target stringlengths 2 27.4k |
|---|---|---|---|
db-write_then_read_k-h30-r0-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:04ede112e26cc0c601bbd6dfc26e4e15c77d6b8bbff6e271a7078b59069b4311",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {
... | [{"branch_owner_sum": 200165, "control_id": 100082, "edge_code": "cmhme5f0", "history_count": 20, "history_edges": "57412448882839835280", "history_weight": 16005, "primary_id": 100081, "weighted_state": 2352}] |
db-radius-L2-r0-radius-L2-precision-3 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:04ede112e26cc0c601bbd6dfc26e4e15c77d6b8bbff6e271a7078b59069b4311",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L2",
"scoring":... | [{"city": "seville", "first_status": "pending", "id": 2, "last_status": "pending", "name": "Alice Kim", "order_count": 1, "order_weight": 8}] |
hc3-game-computation-h3-r506-n0048-2-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring":... | Total score: 106. |
ar1-shell-entity-delete-h3-r30000-n0000-entity_delete_then_use-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
... | audit_root='/work/etc'
slot1_key=chrony/cache.conf
slot1_state=present
slot1_revision=20
slot2_key=collectd/limits.conf
slot2_state=present
slot2_revision=50
slot3_key=memcached/listeners.conf
slot3_state=present
slot3_revision=22
slot4_key=memcached/server.conf
slot4_state=present
slot4_revision=36
live_total=5
audit_... |
hc3-game-computation-h3-r500-n0000-2-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring":... | Total score: 144. |
hc3-game-quota-h3-r504-n0016-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring"... | apple is full; crafting quota exceeded. |
ar2g-game-quota-recovery-h10-r40010-n0041-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring... | carrot is full; crafting quota exceeded. |
game-quota_enforcement_probe-h3-r0-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:723f8e8dae1d98deb7cc38d7ef97ffedcfdf48e2d2260744422abda41a3639d9",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring"... | carrot is full; crafting quota exceeded. |
db-adversarial_persistence-h30-r0-adversarial_persistence-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:e2e30ba2e49e1d4740e29b623fa1446c4f61ee0698b1300fc5f009ffbfd47a29",
"h_reachback": 7,
"item_shape": "output_prediction",
"operator": "adversarial_persistence",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scori... | [{"branch_owner_sum": 200153, "control_id": 100076, "edge_code": "vnhal7w0", "history_count": 15, "history_edges": "736789430708359", "history_weight": 10301, "primary_id": 100075, "weighted_state": 2310}] |
shell-computation_probe-h30-r0-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "shell",
"executor_image": "alpine@sha256:6baf43584bcb78f2e5847d1de515f23499913ac9f12bdf834811a3145eb11ca1",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"... | path=/work/etc/chrony/server.conf lines=1 bytes=3 digest=365
path=/work/etc/collectd/limits.conf lines=1 bytes=3 digest=693
path=/work/etc/redis/cache.conf lines=1 bytes=3 digest=831
|
code-inverse_pair-h3-r0-inverse_pair-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "inverse_pair",
"primary_sub_eval": "irreversibility",
"radius_level": "L0",
"scoring": {
"adju... | ...F [100%]
=========================== short test summary info ============================
FAILED tests/test_retry_backoff.py::test_module_constants - AssertionError: a...
1 failed, 3 passed in Xs
inverse_group_audit selected=ledger/retry_backoff.py ... |
code-transition_correctness_probe-h30-r0-transition_correctness_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "transition_correctness_probe",
"primary_sub_eval": "transition_correctness",
"radius_level": "L0",
... | M ledger/audit_trail.py
|
pb2-api-error-h10-r12001-n0013-GorillaFileSystem-1-error_fidelity_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "error_fidelity_probe",
"primary_sub_eval": "error_fidelity",
"radius_level": "L0",
"scoring"... | FileNotFoundError: wc_many completed=[{"file_name":"registry-quotas.conf","lines":1,"bytes":16},{"file_name":"search-timeouts.conf","lines":3,"bytes":50}]; wc: router-backpressure.conf: No such file or directory |
db-adversarial_persistence-h3-r0-adversarial_persistence-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:e2e30ba2e49e1d4740e29b623fa1446c4f61ee0698b1300fc5f009ffbfd47a29",
"h_reachback": 19,
"item_shape": "output_prediction",
"operator": "adversarial_persistence",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scor... | [{"branch_owner_sum": 200090, "control_id": 100044, "edge_code": "ckmds0t0", "history_count": 15, "history_edges": "633873972109804", "history_weight": 9699, "primary_id": 100045, "weighted_state": 2354}] |
ar2g-game-quota-recovery-h3-r40012-n0048-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:fa3195acdcf4a755c3c756d7a9e3f5cf82db7870d09aab5522ba3705a4a2eb64",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring"... | strawberry is full; crafting quota exceeded. |
api-sycophantic_confirmation_probe-h3-r0-GorillaFileSystem-0-sycophantic_confirmation_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:ef410ddd07d6258c8fbb441b16af1bfd87ef6bc3c60db49590ff764e2bc84ed8",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "sycophantic_confirmation_probe",
"primary_sub_eval": "sycophantic_confirmation",
"radius_level"... | {"audit": {"write_files": ["billing-routing.conf", "billing-cutover.conf", "billing-routing.conf", "billing-cutover.conf", "billing-cutover.conf", "cutover-handoff.conf", "billing-routing.conf", "billing-cutover.conf"], "write_bytes": [54, 48, 46, 45, 72, 154, 75, 70], "transitions": 6}, "target": "billing-cutover.conf... |
hc3-game-quota-h3-r500-n0000-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:7c95de5d0b6f661f2c7a201539882a1f5eb97e6b36233c6b2405783978d87d61",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring"... | peanut is full; crafting quota exceeded. |
pb2-api-error-h3-r12000-n0008-GorillaFileSystem-2-error_fidelity_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "error_fidelity_probe",
"primary_sub_eval": "error_fidelity",
"radius_level": "L0",
"scoring":... | FileNotFoundError: wc_many completed=[{"file_name":"scheduler-warmup.conf","lines":2,"bytes":33},{"file_name":"ledger-routing.conf","lines":1,"bytes":16},{"file_name":"ingest-backpressure.conf","lines":3,"bytes":50}]; wc: ingest-shards.conf: No such file or directory |
lcms1-api-sycophancy-h3-r60000-n0000-GorillaFileSystem-0-sycophantic_confirmation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "sycophantic_confirmation_probe",
"primary_sub_eval": "sycophantic_confirmation",
"radius_level"... | {"audit": {"write_files": ["billing-replicas.conf", "billing-limits.conf", "billing-replicas.conf", "billing-limits.conf", "billing-limits.conf", "cutover-handoff.conf", "billing-replicas.conf", "billing-limits.conf"], "write_bytes": [84, 54, 46, 46, 73, 155, 75, 71], "transitions": 6}, "target": "billing-replicas.conf... |
hc3-db-persistence-h10-r511-n0045-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:04ede112e26cc0c601bbd6dfc26e4e15c77d6b8bbff6e271a7078b59069b4311",
"h_reachback": 3,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {
... | [{"branch_owner_sum": 200079, "control_id": 100039, "edge_code": "tsmau3d0", "history_count": 10, "history_edges": "3147669062", "history_weight": 4837, "primary_id": 100038, "weighted_state": 2276}] |
db-adversarial_persistence-h10-r0-adversarial_persistence-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:e2e30ba2e49e1d4740e29b623fa1446c4f61ee0698b1300fc5f009ffbfd47a29",
"h_reachback": 3,
"item_shape": "output_prediction",
"operator": "adversarial_persistence",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scori... | [{"branch_owner_sum": 200089, "control_id": 100044, "edge_code": "hklst6s1", "history_count": 15, "history_edges": "603160693217137", "history_weight": 9802, "primary_id": 100043, "weighted_state": 2186}] |
api-radius-L3-r3-TwitterAPI-0-radius-L3-precision-4 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L3",
"sc... | {"tweet_id": 14, "exists": true, "content_chars": 54, "tag_count": 1, "mention_count": 1, "comment_count": 0, "retweet_count": 0, "audit_events": [{"operation": "post", "tweet_id": 16, "tweet_count": 8}, {"operation": "post", "tweet_id": 17, "tweet_count": 9}]} |
code-transition_correctness_probe-h10-r2-transition_correctness_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "transition_correctness_probe",
"primary_sub_eval": "transition_correctness",
"radius_level": "L0",
... | M roster/batch_sizing.py
|
code-transition_correctness_probe-h10-r1-transition_correctness_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "transition_correctness_probe",
"primary_sub_eval": "transition_correctness",
"radius_level": "L0",
... | M dispatch/digest_window.py
|
api-radius-L1-r0-TwitterAPI-0-radius-L1-precision-1 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:ef410ddd07d6258c8fbb441b16af1bfd87ef6bc3c60db49590ff764e2bc84ed8",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L1",
"sc... | {"tweet_id": 13, "exists": true, "content_chars": 50, "tag_count": 1, "mention_count": 1, "comment_count": 0, "retweet_count": 0, "audit_events": [{"operation": "mention", "tweet_id": 12, "tweet_count": 7}]} |
db-quota_enforcement_probe-h100-r2-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:04ede112e26cc0c601bbd6dfc26e4e15c77d6b8bbff6e271a7078b59069b4311",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring"... | IntegrityError: CHECK constraint failed: reservation_allowance_exhausted |
hc3-db-persistence-h3-r505-n0020-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:04ede112e26cc0c601bbd6dfc26e4e15c77d6b8bbff6e271a7078b59069b4311",
"h_reachback": 1,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {
... | [{"branch_owner_sum": 200141, "control_id": 100070, "edge_code": "tkjri5i9", "history_count": 10, "history_edges": "5817188994", "history_weight": 5121, "primary_id": 100069, "weighted_state": 2355}] |
hc3-code-persistence-h10-r512-n0049-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 2,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {... | wm_code_group_audit current=865 cross=966 events=16 transitions=14 digest=761
|
hc1-db-access-h3-r8-n0012-access_control_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:04ede112e26cc0c601bbd6dfc26e4e15c77d6b8bbff6e271a7078b59069b4311",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "access_control_probe",
"primary_sub_eval": "access_control",
"radius_level": "L0",
"scoring": {
... | IntegrityError: CHECK constraint failed: delete_blocked_allocation_open |
game-radius-L1-r0-radius-L2-precision-2 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:723f8e8dae1d98deb7cc38d7ef97ffedcfdf48e2d2260744422abda41a3639d9",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L2",
"scoring... | The mantelpiece is solid. |
math-radius-L3-r2-radius-L3-recall-1 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "math_and_science",
"domain_legacy": "math",
"executor_image": "sha256:c090e8bf022ef82adc34b5474e8ddda68472632e3828e8cca69a884056c8a745",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L3"... | ['radius_state_audit', 174987/484, -2025/44, False, 21, 684645] |
math-radius-L3-r0-radius-L3-recall-1 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "math_and_science",
"domain_legacy": "math",
"executor_image": "sha256:d28e201a98758cc35fe4a0300520f92192a3d3476f9703c72460935508cc2e3f",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L3"... | ['radius_state_audit', 4082665/24, 11407/18, False, 21, 167067] |
code-write_then_read_k-h10-r1-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 1,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {... | wm_code_group_audit current=150 cross=829 events=24 transitions=22 digest=97
|
lh-math-k0-write_read_persistence-h600-state_injection | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "math_and_science",
"domain_legacy": "math",
"executor_image": "sha256:c090e8bf022ef82adc34b5474e8ddda68472632e3828e8cca69a884056c8a745",
"h_reachback": 600,
"item_shape": "output_prediction",
"operator": "horizon_family",
"primary_sub_eval": "write_read_persistence",
"radius_level": null,
... | ['wm_group_audit', 71*(x - 9)*(x + 9) + 59*(x - 8)*(x + 5) + 97*(x - 8)*(x + 7) + 43*(x - 7)*(x + 4) + 17*(x - 7)*(x + 6) + 83*(x - 6)*(x + 2) + 31*(x - 4)*(x + 4) + 127*(x - 2)*(x + 3) + 109*(x - 2)*(x + 5) + 149*(x - 2)*(x + 9), (x - 9)**2*(x + 9)**2 - (x - 8)**2*(x + 5)**2 + (x - 8)**2*(x + 7)**2 + (x - 7)**2*(x + 4... |
api-near_twin_mutate-h100-r1-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scorin... | FileNotFoundError: configuration audit {'write_files': ['registry-warmup.conf', 'gateway-warmup.conf', 'router-limits.conf', 'billing-quotas.conf', 'registry-warmup.conf', 'router-limits.conf', 'gateway-warmup.conf', 'billing-quotas.conf', 'gateway-warmup.conf', 'registry-warmup.conf', 'billing-quotas.conf', 'router-li... |
code-inverse_pair-h100-r0-inverse_pair-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "inverse_pair",
"primary_sub_eval": "irreversibility",
"radius_level": "L0",
"scoring": {
"ad... | ...F [100%]
=========================== short test summary info ============================
FAILED tests/test_digest_window.py::test_module_constants - AssertionError: a...
1 failed, 3 passed in Xs
inverse_group_audit selected=dispatch/digest_window.p... |
api-access_control_probe-h3-r0-TicketAPI-1-access_control_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "access_control_probe",
"primary_sub_eval": "access_control",
"radius_level": "L0",
"scoring":... | [{"error": "User not authenticated. Please log in to view tickets."}] |
code-write_then_read-h10-r1-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 2,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {... | wm_code_group_audit current=881 cross=381 events=16 transitions=14 digest=874
|
dialogue-temporal_ordering_probe-h100-r0-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "dialogue",
"domain_legacy": "dialogue",
"executor_image": "sha256:270c7b5b9ea0629def553706c617a7e1373297bb550bd3c8f08b1b2ab1a60f7a",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L0",
... | {"events": [{"key": "#W7538230", "kind": "order", "seq": 2, "value": "queued"}, {"key": "#W2624389", "kind": "number", "seq": 3, "value": 7}, {"key": "#W1890669", "kind": "customer", "seq": 4, "value": "queued"}, {"key": "#W2806889", "kind": "string", "seq": 5, "value": "7"}, {"key": "#W2624389", "kind": "order", "seq"... |
api-near_twin_mutate-h100-r0-GorillaFileSystem-0-near_twin_mutate-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scorin... | FileNotFoundError: configuration audit {'write_files': ['registry-routing.conf', 'indexer-quotas.conf', 'router-quotas.conf', 'gateway-warmup.conf', 'registry-routing.conf', 'router-quotas.conf', 'indexer-quotas.conf', 'gateway-warmup.conf', 'indexer-quotas.conf', 'registry-routing.conf', 'gateway-warmup.conf', 'router... |
code-format_stability_probe-h100-r0-format_stability_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scoring"... | [index-numstat]
[worktree-numstat]
3 1 billing/cache_expiry.py
4 2 billing/rate_limit.py
1 1 billing/retry_backoff.py
1 1 billing/worker_pool.py
[selected-index]
[selected-worktree]
diff --git a/billing/cache_expiry.py b/billing/cache_expiry.py
--- a/billing/cache_expiry.py
+++ b/billing/cache_expiry.py
@@ -3,3 +3,3 @@... |
code-transition_correctness_probe-h100-r0-transition_correctness_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "transition_correctness_probe",
"primary_sub_eval": "transition_correctness",
"radius_level": "L0",... | M ledger/session_window.py
|
lh-db-k4-write_read_persistence-h300-state_injection | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "db",
"executor_image": "sha256:04ede112e26cc0c601bbd6dfc26e4e15c77d6b8bbff6e271a7078b59069b4311",
"h_reachback": 300,
"item_shape": "output_prediction",
"operator": "horizon_family",
"primary_sub_eval": "write_read_persistence",
"radius_level": null,
"scoring": {
... | [{"alternating_state": 21, "extrema_sum": 90, "ordered_edges": "J3VnPoMsCsPnHyAsCoAo", "square_state": 19271, "transitions": "-8,-4,1,3,-6,5,1,1,-9", "weighted_state": 2345}] |
api-radius-L1-r0-GorillaFileSystem-0-radius-L1-precision-2 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:ef410ddd07d6258c8fbb441b16af1bfd87ef6bc3c60db49590ff764e2bc84ed8",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L1",
"sc... | {"target": {"file_content": "umber-juniper-51"}, "namespace": {"current_directory_content": ["archive", "capacity.csv", "runbook.md", "ingest-shards.conf", "ledger-quotas.conf", "ingest-quotas.conf", "scheduler-shards.conf", "ledger-1.logs", "registry-2.logs"]}, "ordered_writes": [{"file_name": "ingest-shards.conf", "b... |
api-format_stability_probe-h3-r0-GorillaFileSystem-0-format_stability_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:ef410ddd07d6258c8fbb441b16af1bfd87ef6bc3c60db49590ff764e2bc84ed8",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scori... | {"files": [{"file_name": "gateway-warmup.conf", "canonical_content": "# interleaved revision ledger\nlog_level=info\nqueue_depth=128\nrevision_token=meridian-56\nroute_weight=14\nsync_epoch=6"}, {"file_name": "registry-backpressure.conf", "canonical_content": "# interleaved revision ledger\nlog_level=info\nqueue_depth=... |
api-format_stability_probe-h10-r0-GorillaFileSystem-0-format_stability_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:ef410ddd07d6258c8fbb441b16af1bfd87ef6bc3c60db49590ff764e2bc84ed8",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scor... | {"files": [{"file_name": "registry-backpressure.conf", "canonical_content": "# interleaved revision ledger\nlog_level=info\nqueue_depth=256\nrevision_token=cedar-38\nroute_weight=37\nsync_epoch=6"}, {"file_name": "scheduler-cutover.conf", "canonical_content": "# interleaved revision ledger\nlog_level=info\nqueue_depth=... |
game-transition_correctness_probe-h30-r0-transition_correctness_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "game",
"domain_legacy": "game",
"executor_image": "sha256:723f8e8dae1d98deb7cc38d7ef97ffedcfdf48e2d2260744422abda41a3639d9",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "transition_correctness_probe",
"primary_sub_eval": "transition_correctness",
"radius_level": "L0",
... | The cabinet looks strong, and impossible to break. It is open.
In the cabinet is a tv. |
dialogue-radius-L1-r1-radius-L1-precision-1 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "dialogue",
"domain_legacy": "dialogue",
"executor_image": "sha256:270c7b5b9ea0629def553706c617a7e1373297bb550bd3c8f08b1b2ab1a60f7a",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L1",
... | {"focus": {"entity": "users.aarav_sanchez_6568", "kind": "user", "order_count": 0, "payment_method_count": 1}, "records": [{"entity": "users.aarav_sanchez_6568", "kind": "user", "order_count": 0, "payment_method_count": 1}, {"entity": "users.sofia_li_3261", "kind": "user", "order_count": 2, "payment_method_count": 1}, ... |
api-computation_probe-h100-r0-GorillaFileSystem-0-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"sc... | {"diff_lines": "- 12\n+ 92"} |
api-content_repetition_probe-h100-r0-GorillaFileSystem-0-content_repetition_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "content_repetition_probe",
"primary_sub_eval": "content_repetition",
"radius_level": "L0",
... | {"records": [{"file_name": "billing-routing.conf", "file_content": "log_level=warn\nretry_limit=2\nqueue_depth=512", "bytes": 44}, {"file_name": "registry-retention.conf", "file_content": "log_level=warn\nretry_limit=3\nqueue_depth=1024", "bytes": 45}, {"file_name": "ledger-quotas.conf", "file_content": "log_level=warn... |
api-content_repetition_probe-h3-r1-GorillaFileSystem-0-content_repetition_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "content_repetition_probe",
"primary_sub_eval": "content_repetition",
"radius_level": "L0",
"s... | {"records": [{"file_name": "gateway-replicas.conf", "file_content": "log_level=warn\nretry_limit=3\nqueue_depth=512", "bytes": 44}, {"file_name": "scheduler-limits.conf", "file_content": "log_level=warn\nretry_limit=4\nqueue_depth=512", "bytes": 44}, {"file_name": "notifier-cutover.conf", "file_content": "log_level=war... |
api-dual_path_read-h3-r2-GorillaFileSystem-0-dual_path_read-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "dual_path_read",
"primary_sub_eval": "contradiction_detection",
"radius_level": "L0",
"scorin... | {"files": [{"file_name": "ledger-timeouts.conf", "file_content": "fjord-sable-59"}]} |
api-error_fidelity_probe-h30-r1-GorillaFileSystem-0-error_fidelity_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "error_fidelity_probe",
"primary_sub_eval": "error_fidelity",
"radius_level": "L0",
"scoring"... | {"files": [{"file_name": "ledger-timeouts.conf", "lines": 1, "bytes": 16}, {"file_name": "indexer-quotas.conf", "lines": 1, "bytes": 16}, {"file_name": "registry-limits.conf", "lines": 3, "bytes": 50}, {"file_name": "search-routing.conf", "lines": 2, "bytes": 33}]} |
api-format_stability_probe-h100-r1-GorillaFileSystem-0-format_stability_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"sco... | {"files": [{"file_name": "notifier-routing.conf", "canonical_content": "# interleaved revision ledger\nlog_level=info\nqueue_depth=128\nrevision_token=fjord-22\nroute_weight=38\nsync_epoch=8"}, {"file_name": "notifier-retention.conf", "canonical_content": "# interleaved revision ledger\nlog_level=notice\nqueue_depth=12... |
api-format_stability_probe-h3-r1-GorillaFileSystem-0-format_stability_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "format_stability_probe",
"primary_sub_eval": "format_stability",
"radius_level": "L0",
"scori... | {"files": [{"file_name": "gateway-shards.conf", "canonical_content": "# interleaved revision ledger\nlog_level=notice\nqueue_depth=1024\nrevision_token=basalt-50\nroute_weight=6\nsync_epoch=8"}, {"file_name": "search-replicas.conf", "canonical_content": "# interleaved revision ledger\nlog_level=info\nqueue_depth=256\nr... |
api-inverse_pair-h100-r1-GorillaFileSystem-0-inverse_pair-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 105,
"item_shape": "output_prediction",
"operator": "inverse_pair",
"primary_sub_eval": "irreversibility",
"radius_level": "L0",
"scoring": {
... | {"target": "search-routing.conf", "target_bytes": 18, "file_content": "juniper-krypton-27", "audit": {"write_files": ["search-routing.conf", "search-backpressure.conf", "billing-retention.conf", "ledger-shards.conf", "search-routing.conf", "search-backpressure.conf", "billing-retention.conf", "ledger-shards.conf"], "wr... |
api-partial_failure_probe-h30-r1-TicketAPI-0-partial_failure_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "partial_failure_probe",
"primary_sub_eval": "partial_failure_semantics",
"radius_level": "L0",... | {"error": "Invalid fields for update: created_by"} |
api-quota_enforcement_probe-h100-r0-GorillaFileSystem-0-quota_enforcement_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"s... | {"error": "operation quota exceeded", "resource": "indexer-routing.conf", "charge": 15, "limit": 135, "used": 134, "requested": 149} |
api-referential_integrity_probe-h30-r0-GorillaFileSystem-0-referential_integrity_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "referential_integrity_probe",
"primary_sub_eval": "referential_integrity",
"radius_level": "L0... | {"reference": "control.ref", "target": "notifier-backpressure.conf", "file_content": "lumen-umber-56"} |
api-replay_determinism_probe-h3-r1-GorillaFileSystem-0-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"s... | {"files": [{"file_name": "ledger-cutover.conf", "file_content": "krypton-garnet-66"}, {"file_name": "gateway-retention.conf", "file_content": "garnet-fjord-43"}]} |
api-redundant_action-h3-r0-GorillaFileSystem-1-redundant_action-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 22,
"item_shape": "output_prediction",
"operator": "redundant_action",
"primary_sub_eval": "rule_enforcement",
"radius_level": "L3",
"scoring": ... | {"error": "", "audit": {"records": [{"file_name": "search-quotas.conf", "exists": true, "bytes": 14}, {"file_name": "billing-retention.conf", "exists": true, "bytes": 16}, {"file_name": "ledger-timeouts.conf", "exists": true, "bytes": 16}, {"file_name": "ledger-warmup.conf", "exists": false, "bytes": 0}, {"file_name": ... |
api-schema_conformance_probe-h30-r1-GorillaFileSystem-0-schema_conformance_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "schema_conformance_probe",
"primary_sub_eval": "schema_conformance",
"radius_level": "L0",
"... | {"matching_lines": ["queue_depth=1024", "log_level=info", "idle_timeout_s=600"]} |
api-radius-L1-r0-GorillaFileSystem-0-radius-L1-precision-1 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L1",
"sc... | {"target": {"file_content": "onyx-beacon-37"}, "namespace": {"current_directory_content": ["archive", "capacity.csv", "runbook.md", "ingest-shards.conf", "ledger-quotas.conf", "ingest-quotas.conf", "scheduler-shards.conf", "ledger-1.logs", "registry-2.logs"]}, "ordered_writes": [{"file_name": "ingest-shards.conf", "byt... |
api-radius-L1-r0-GorillaFileSystem-0-radius-L1-recall-0 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L1",
"sc... | {"target": {"file_content": "willow-lumen-93"}, "namespace": {"current_directory_content": ["archive", "capacity.csv", "runbook.md", "ingest-shards.conf", "ledger-quotas.conf", "ingest-quotas.conf", "scheduler-shards.conf", "ledger-1.logs", "registry-2.logs"]}, "ordered_writes": [{"file_name": "ingest-shards.conf", "by... |
api-radius-L1-r1-GorillaFileSystem-0-radius-L1-recall-0 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L1",
"sc... | {"target": {"file_content": "fjord-delta-53"}, "namespace": {"current_directory_content": ["archive", "capacity.csv", "runbook.md", "registry-quotas.conf", "ingest-routing.conf", "billing-shards.conf", "search-quotas.conf", "indexer-1.logs", "search-2.logs"]}, "ordered_writes": [{"file_name": "registry-quotas.conf", "b... |
api-radius-L1-r1-TicketAPI-0-radius-L1-precision-1 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L1",
"sc... | {"ticket_id": 42, "exists": true, "field_count": 6, "audit_events": [{"operation": "create", "ticket_id": 41, "ticket_count": 4}, {"operation": "create", "ticket_id": 42, "ticket_count": 5}, {"operation": "create", "ticket_id": 43, "ticket_count": 6}, {"operation": "edit", "ticket_id": 41, "ticket_count": 6}, {"operati... |
api-radius-L2-r0-GorillaFileSystem-0-radius-L2-recall-1 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L2",
"sc... | {"matches": ["./archive/ingest-cutover.conf", "./ingest-cutover.conf"]} |
api-radius-L2-r1-GorillaFileSystem-0-radius-L2-precision-4 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L2",
"sc... | {"target": {"file_content": "vector-beacon-74"}, "namespace": {"current_directory_content": ["archive", "capacity.csv", "runbook.md", "ingest-quotas.conf", "ledger-warmup.conf", "registry-quotas.conf", "search-cutover.conf", "indexer-1.logs", "gateway-2.logs"]}, "ordered_writes": [{"file_name": "ingest-quotas.conf", "b... |
api-sycophantic_confirmation_probe-h100-r0-GorillaFileSystem-0-sycophantic_confirmation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "sycophantic_confirmation_probe",
"primary_sub_eval": "sycophantic_confirmation",
"radius_leve... | {"audit": {"write_files": ["router-replicas.conf", "gateway-quotas.conf", "router-replicas.conf", "gateway-quotas.conf", "gateway-quotas.conf", "cutover-handoff.conf", "router-replicas.conf", "gateway-quotas.conf"], "write_bytes": [63, 81, 47, 46, 73, 154, 76, 71], "transitions": 6}, "target": "router-replicas.conf", "... |
api-temporal_ordering_probe-h100-r0-GorillaFileSystem-0-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L0",
"s... | {"matches": ["./scheduler-shards.conf", "./billing-quotas.conf", "./router-warmup.conf"]} |
api-transition_correctness_probe-h3-r1-TicketAPI-0-transition_correctness_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "transition_correctness_probe",
"primary_sub_eval": "transition_correctness",
"radius_level": "L... | {"id": 41, "title": "Webhook relay: slow startup", "description": "found while draining the replay backlog", "status": "Closed", "priority": 5, "created_by": "j.meyer", "resolution": "on-call paged after the nightly rebalance; rolled forward after the warmup window."} |
api-write_then_read-h10-r0-GorillaFileSystem-0-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 2,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scorin... | {"audit": {"write_files": ["indexer-warmup.conf", "router-limits.conf", "registry-backpressure.conf", "billing-shards.conf", "indexer-warmup.conf", "registry-backpressure.conf", "billing-shards.conf", "router-limits.conf", "router-limits.conf", "registry-backpressure.conf", "router-limits.conf", "billing-shards.conf", ... |
api-write_then_read-h100-r0-GorillaFileSystem-0-write_then_read-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 114,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scor... | {"audit": {"write_files": ["gateway-backpressure.conf", "registry-backpressure.conf", "gateway-replicas.conf", "ingest-backpressure.conf", "gateway-backpressure.conf", "gateway-replicas.conf", "ingest-backpressure.conf", "registry-backpressure.conf", "registry-backpressure.conf", "gateway-replicas.conf", "registry-back... |
api-write_then_read-h3-r0-GorillaFileSystem-0-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 1,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scorin... | {"audit": {"write_files": ["billing-backpressure.conf", "ingest-warmup.conf", "notifier-retention.conf", "router-timeouts.conf", "ingest-warmup.conf", "notifier-retention.conf", "router-timeouts.conf", "billing-backpressure.conf", "billing-backpressure.conf", "notifier-retention.conf", "billing-backpressure.conf", "rou... |
api-write_then_read-h3-r0-TicketAPI-0-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "enterprise",
"domain_legacy": "api",
"executor_image": "sha256:4bc39747926d4eaff42ada2c81ef489e52fb836f4217af5a6d342f66b3def06f",
"h_reachback": 1,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scorin... | {"records": [{"ticket_id": 41, "description": "harbor-lumen-45; relay audit 03, pass 15.", "status": "Open", "priority": 3}, {"ticket_id": 42, "description": "fjord-willow-41; relay audit 58, pass 1.", "status": "Open", "priority": 3}, {"ticket_id": 43, "description": "meridian-sable-13; relay audit 35, pass 14.", "sta... |
code-access_control_probe-h100-r0-access_control_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "access_control_probe",
"primary_sub_eval": "access_control",
"radius_level": "L0",
"scoring": {
... | bash: line 1: ./scripts/sync_shards.sh: Permission denied
|
code-computation_probe-h10-r1-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring"... | 16 catalog/session_window.py
12 catalog/retry_backoff.py
16 catalog/queue_drain.py
44 total
|
code-computation_probe-h100-r0-computation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "computation_probe",
"primary_sub_eval": "computational_accuracy",
"radius_level": "L0",
"scoring... | 12 ingest/shard_map.py
8 ingest/payload_limit.py
12 ingest/batch_sizing.py
32 total
|
code-content_repetition_probe-h100-r1-content_repetition_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "content_repetition_probe",
"primary_sub_eval": "content_repetition",
"radius_level": "L0",
"scor... | module=warehouse/cache_expiry.py tag=juniper-cove-39 bytes=234
module=warehouse/fee_rounding.py tag=delta-cove-08 bytes=167
module=warehouse/rate_limit.py tag=amber-beacon-58 bytes=180
module=warehouse/retry_backoff.py tag=basalt-isthmus-25 bytes=263
module=warehouse/shard_map.py tag=lumen-lantern-17 bytes=244
module=w... |
code-near_twin_mutate-h100-r0-near_twin_mutate-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "near_twin_mutate",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scoring": {... | """Queue drain budgeting for the catalog service."""
"""Queue drain budgeting for the catalog service."""
PROFILE_TAG = "krypton-onyx-26"
DRAIN_CHUNK = 64
def drain_budget(workers):
return workers * DRAIN_CHUNK
def aggregate_drain_budget(values):
return sum(drain_budget(v) for v in values)
def peak_drain... |
code-dual_path_read-h100-r1-dual_path_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "dual_path_read",
"primary_sub_eval": "contradiction_detection",
"radius_level": "L0",
"scoring":... | selected=cedar-garnet-63 worktree=cedar-garnet-63 committed=lumen-cove-51 expected=delta-kiln-07 relation=diverged
|
code-error_fidelity_probe-h30-r0-error_fidelity_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "error_fidelity_probe",
"primary_sub_eval": "error_fidelity",
"radius_level": "L0",
"scoring": {
... | neighbor=sable-juniper-34
Traceback (most recent call last):
File "<string>", line 1, in <module>
ImportError: cannot import name 'PROFILE_TAG' from 'warehouse.batch_sizing' (/repo/warehouse/batch_sizing.py)
|
code-partial_failure_probe-h10-r1-partial_failure_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 11,
"item_shape": "output_prediction",
"operator": "partial_failure_probe",
"primary_sub_eval": "partial_failure_semantics",
"radius_level": "L2",
"s... | rc=0
roster/digest_window.py:"""Digest windowing for the roster platform."""
roster/shard_map.py:"""Shard assignment for the roster platform."""
roster/rate_limit.py:"""Rate-limit window budgeting for the roster platform."""
|
code-quota_enforcement_probe-h3-r2-quota_enforcement_probe-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "quota_enforcement_probe",
"primary_sub_eval": "quota_enforcement",
"radius_level": "L0",
"scoring"... | preflight target=roster/audit_trail.py dependency=roster/queue_drain.py remaining=1
accepted=1 remaining_after=0
|
code-referential_integrity_probe-h3-r0-referential_integrity_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 4,
"item_shape": "output_prediction",
"operator": "referential_integrity_probe",
"primary_sub_eval": "referential_integrity",
"radius_level": "L0",
"... |
==================================== ERRORS ====================================
_________________ ERROR collecting tests/test_retry_backoff.py _________________
ImportError while importing test module '/repo/tests/test_retry_backoff.py'.
Hint: make sure your test modules/packages have valid Python names.
Traceback:
/... |
code-replay_determinism_probe-h100-r1-replay_determinism_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "replay_determinism_probe",
"primary_sub_eval": "replay_determinism",
"radius_level": "L0",
"scor... | """Audit trail capacity for the telemetry service."""
PROFILE_TAG = "cedar-vector-65"
TRAIL_LIMIT = 120
def trail_capacity(days):
return days * TRAIL_LIMIT
def aggregate_trail_capacity(values):
return sum(trail_capacity(v) for v in values)
# capacity checkpoint 1: profile harbor-flint-54
# capacity check... |
code-schema_conformance_probe-h100-r0-schema_conformance_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "schema_conformance_probe",
"primary_sub_eval": "schema_conformance",
"radius_level": "L0",
"scor... | Expecting ',' delimiter: line 6 column 3 (char 104)
|
code-radius-L3-r0-radius-L3-precision-3 | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": null,
"item_shape": "post_action_probe",
"operator": "escalate_radius",
"primary_sub_eval": "side_effect_propagation",
"radius_level": "L3",
"scoring... | module_rollup selected=git:ingest/rate_limit.py records=[ingest/rate_limit.py|ember-krypton-71|service|30;ingest/shard_map.py|lumen-marsh-55|service|16;ingest/fee_rounding.py|ember-beacon-44|service|40;ingest/retry_backoff.py|basalt-estuary-79|service|40;ingest/digest_window.py|cedar-grove-65|service|6;] platform_count... |
code-sycophantic_confirmation_probe-h100-r0-sycophantic_confirmation_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "sycophantic_confirmation_probe",
"primary_sub_eval": "sycophantic_confirmation",
"radius_level": "... | module=ingest/cache_expiry.py profile_tag=lumen-lantern-06 bytes=232
module=ingest/digest_window.py profile_tag=meridian-quarry-94 bytes=321
module=ingest/payload_limit.py profile_tag=ember-jetty-87 bytes=258
module=ingest/rate_limit.py profile_tag=harbor-hollow-74 bytes=266
module=ingest/retry_backoff.py profile_tag=l... |
code-temporal_ordering_probe-h100-r0-temporal_ordering_probe-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 101,
"item_shape": "output_prediction",
"operator": "temporal_ordering_probe",
"primary_sub_eval": "temporal_ordering",
"radius_level": "L0",
"scorin... | roll retry backoff schedule back to the previous profile
pin retry backoff schedule to the reviewed profile
tune retry backoff schedule for the release
add retry backoff schedule module
|
code-entity_delete_then_use-h10-r0-entity_delete_then_use-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 14,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scorin... | committed=catalog/retry_backoff.py lines=1 defs=0
committed=catalog/worker_pool.py lines=12 defs=2
committed=tests/test_retry_backoff.py lines=1 defs=0
committed=tests/test_worker_pool.py lines=19 defs=4
from catalog.worker_pool import POOL_WORKERS, PROFILE_TAG, pool_capacity
def test_profile_tag():
assert PROFIL... |
code-entity_delete_then_use-h100-r0-entity_delete_then_use-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 104,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scori... | committed=dispatch/audit_trail.py lines=12 defs=2
committed=dispatch/batch_sizing.py lines=1 defs=0
committed=tests/test_audit_trail.py lines=19 defs=4
committed=tests/test_batch_sizing.py lines=1 defs=0
from dispatch.audit_trail import TRAIL_LIMIT, PROFILE_TAG, trail_capacity
def test_profile_tag():
assert PROFI... |
code-entity_delete_then_use-h100-r1-entity_delete_then_use-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 104,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scori... | committed=warehouse/audit_trail.py lines=12 defs=2
committed=warehouse/fee_rounding.py lines=1 defs=0
committed=tests/test_audit_trail.py lines=19 defs=4
committed=tests/test_fee_rounding.py lines=1 defs=0
from warehouse.audit_trail import TRAIL_LIMIT, PROFILE_TAG, trail_capacity
def test_profile_tag():
assert PR... |
code-entity_delete_then_use-h30-r0-entity_delete_then_use-twin | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 34,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scorin... | committed=dispatch/queue_drain.py lines=16 defs=3
committed=dispatch/retry_backoff.py lines=1 defs=0
committed=tests/test_queue_drain.py lines=19 defs=4
committed=tests/test_retry_backoff.py lines=1 defs=0
from dispatch.queue_drain import DRAIN_CHUNK, PROFILE_TAG, drain_budget
def test_profile_tag():
assert PROFI... |
code-entity_delete_then_use-h30-r1-entity_delete_then_use-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 31,
"item_shape": "output_prediction",
"operator": "entity_delete_then_use",
"primary_sub_eval": "under_over_erroring",
"radius_level": "L0",
"scorin... | fatal: path 'ledger/worker_pool.py' does not exist in 'HEAD'
fatal: path 'tests/test_worker_pool.py' does not exist in 'HEAD'
|
code-write_then_read-h10-r0-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 2,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {... | wm_code_group_audit current=616 cross=844 events=16 transitions=14 digest=148
|
code-write_then_read-h10-r2-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 2,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {... | wm_code_group_audit current=10 cross=655 events=16 transitions=14 digest=267
|
code-write_then_read-h3-r0-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 1,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {... | wm_code_group_audit current=648 cross=642 events=16 transitions=14 digest=403
|
code-write_then_read-h3-r1-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 1,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {... | wm_code_group_audit current=968 cross=554 events=16 transitions=14 digest=470
|
code-write_then_read-h3-r2-write_then_read-injected | [
{
"content": "You are a world model that simulates the environment for an interface agent. For each agent action, output exactly what the real environment would return: the tool output, error message, or observation. Never fabricate a successful result for a call that must fail, and never return an error for a ... | {
"domain": "code",
"domain_legacy": "code",
"executor_image": "sha256:b42569e53ae2838d2f8f6a475e3e7f9b1cdae04b3802e31441c6e02359f95833",
"h_reachback": 1,
"item_shape": "output_prediction",
"operator": "write_then_read",
"primary_sub_eval": "write_read_persistence",
"radius_level": "L0",
"scoring": {... | wm_code_group_audit current=123 cross=637 events=16 transitions=14 digest=456
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.