text
stringlengths
185
73.3k
repo
stringlengths
7
100
path
stringlengths
4
146
language
stringclasses
7 values
hash
stringlengths
16
16
score
float64
7
8.5
stars
int64
0
237k
""" Desktop notifications via notify-send (libnotify). Falls back to QSystemTrayIcon.showMessage() if notify-send is unavailable or reports an error. notify-send works on KDE Plasma (via plasma-integration) and Hyprland (via dunst, mako, or any notification daemon). """ import logging import subprocess import threadi...
RiDDiX/adguard-tray
adguard_tray/notifications.py
.py
176bee79bf3c4975
7.6
15
"""Report whether browsers can bypass AdGuard over HTTP/3 (QUIC). adguard-cli does handle QUIC, but only in `auto` proxy mode: its iptables rules redirect UDP as well as TCP, and `https_filtering.http3_filtering_enabled` decides whether an intercepted QUIC connection is filtered (true) or blocked so the client falls b...
RiDDiX/adguard-tray
adguard_tray/quic.py
.py
abcf8cf1ccf69f16
7.6
15
#!/usr/bin/env python3 """ Test script for the Quantum Hardware MCP HTTP server. Usage: # Start the server in another terminal: python server.py --transport http # Then run this test: python test_http_server.py """ import requests import json import sys import time SERVER_URL = "http://127.0.0.1...
Lokesh-2025/quantum-hardware-mcp
test_http_server.py
.py
cecf6cd481df8809
7.92
6
""" Tests for _find_connected_qubit_subset (server.py), added 2026-08-25 — fixes a real bug reported by a real user (quantum-chemistry-vqe): best_qubits used to pick the top-n qubits by individual score alone, only warning after the fact if they happened not to be connected. Confirmed live and reproducible against real...
Lokesh-2025/quantum-hardware-mcp
tests/test_best_qubits_connectivity.py
.py
778a06968caab4be
7.92
6
from __future__ import annotations from typing import Any from a2a.client.client import ClientCallContext from a2a.types import Role, SendMessageRequest from common.logging_utils import format_payload_log, format_stage_log, summarize_text from common.sse_event_consumer import normalize_event from client_example.scen...
project-openan/a2a-t-sdk-python
a2a-t-sample/subscribe-incident/src/client_example/client_flow.py
.py
94b86fae849c886e
7.66
20
"""Sample scenario input data for the subscribe-incident e2e flow. Contains the scenario tag, agent_card_query, subscription_filter, and diagnosis_context used by the sample client. The prompt generation input is selected based on the configured language (zh-CN / en-US), so the input language matches `A2AT_LANGUAGE`. ...
project-openan/a2a-t-sdk-python
a2a-t-sample/subscribe-incident/src/client_example/scenario_data.py
.py
9eeeac299340aa38
7.66
20
"""Adapter helpers for building A2A protobuf messages (artifacts, status updates, messages).""" from __future__ import annotations import uuid from collections.abc import Mapping from a2a.server.agent_execution.context import RequestContext from a2a.server.events.event_queue import EventQueue from a2a.types import (...
project-openan/a2a-t-sdk-python
a2a-t-sample/subscribe-incident/src/common/a2a_adapter.py
.py
e6016b11964921cd
7.66
20
"""LLM request/response logger with mock fallback support. Patches OpenAIClient.structured and _build_structured_payload to print request/response payloads. When mock is enabled, returns canned responses instead of calling the real LLM API. """ from __future__ import annotations from typing import Any from a2a_t.ll...
project-openan/a2a-t-sdk-python
a2a-t-sample/subscribe-incident/src/common/llm_logger.py
.py
581b0556036a1371
7.66
20
"""Mock LLM data provider for e2e tests when no real API key is available. Only patches DotEnvConfigSource.load to inject a placeholder api_key (so A2ATClient construction does not fail) and exposes is_mock_enabled() / get_mock_response() for llm_logger to call. Mock response files live under resources/mock_responses...
project-openan/a2a-t-sdk-python
a2a-t-sample/subscribe-incident/src/common/mock_llm.py
.py
833f6e0a78b0cbf6
7.66
20
from __future__ import annotations import asyncio import json from collections.abc import Awaitable, Callable, Mapping from a2a.server.agent_execution.context import RequestContext from a2a.server.events.event_queue import EventQueue from a2a.types import Task, TaskArtifactUpdateEvent, TaskState from common.a2a_adapt...
project-openan/a2a-t-sdk-python
a2a-t-sample/subscribe-incident/src/server_example/server_flow.py
.py
a38019456bd39597
7.66
20
from __future__ import annotations import asyncio import os from contextlib import contextmanager from pathlib import Path from typing import Any import uvicorn from a2a.server.agent_execution.agent_executor import AgentExecutor from a2a.server.agent_execution.context import RequestContext from a2a.server.events.even...
project-openan/a2a-t-sdk-python
a2a-t-sample/subscribe-incident/src/server_example/server_main.py
.py
453ed91760bbaf37
7.66
20
from __future__ import annotations from pathlib import Path from typing import Any from a2a_t.config.models import A2ATConfig from a2a_t.llm.config_loader import LLMConfigLoader from a2a_t.llm.factory import LLMClientFactory from a2a_t.negotiation.common.models import ContinueNegotiationInput, StartNegotiationInput ...
project-openan/a2a-t-sdk-python
src/a2a_t/client/a2at_client.py
.py
9d10ee8a6dbe67a3
7.66
20
from __future__ import annotations from typing import Any from a2a_t.negotiation.common.enums import NegotiationRole from a2a_t.negotiation.runtime.base_negotiation_orchestrator import BaseNegotiationOrchestrator class NegotiationOrchestrator(BaseNegotiationOrchestrator): """Bind the shared negotiation runtime ...
project-openan/a2a-t-sdk-python
src/a2a_t/client/negotiation/negotiation_orchestrator.py
.py
6e8f4a9b0409148a
7.66
20
from __future__ import annotations from pathlib import Path from typing import Any from a2a_t.negotiation.common.enums import NegotiationType from a2a_t.negotiation.handling import NegotiationHandler from a2a_t.negotiation.rendering import NegotiationPromptRenderer from a2a_t.negotiation.store import NegotiationState...
project-openan/a2a-t-sdk-python
src/a2a_t/client/negotiation/negotiation_orchestrator_builder.py
.py
d3fd7b9656a0ef72
7.66
20
from __future__ import annotations import json from .models import NormalizedInput class InputNormalizer: """Normalize raw prompt-generation input into a canonical string payload.""" def normalize(self, user_input: str | dict[str, object]) -> NormalizedInput: """Classify user input and convert it i...
project-openan/a2a-t-sdk-python
src/a2a_t/client/prompt_generation/input_normalizer.py
.py
c58fe22ea0569827
7.66
20
from __future__ import annotations from dataclasses import asdict, dataclass @dataclass(slots=True) class NormalizedInput: """Carry the normalized text and the input shape detected upstream.""" input_kind: str normalized_input: str @dataclass(slots=True) class PromptGenerationFailure: """Describe ...
project-openan/a2a-t-sdk-python
src/a2a_t/client/prompt_generation/models.py
.py
d5ad04649d0d003c
7.66
20
from __future__ import annotations import logging from typing import Any from a2a_t.common.prompt_resources import ( PromptResourceNotFoundError, PromptResourceParseError, ) from a2a_t.config.models import PromptRuntimeConfig from a2a_t.prompt.analysis import ScenarioResolutionOrchestrator from a2a_t.prompt.a...
project-openan/a2a-t-sdk-python
src/a2a_t/client/prompt_generation/prompt_generation_orchestrator.py
.py
7e537d1860e11fe1
7.66
20
from __future__ import annotations from typing import Any from a2a_t.common.prompt_runtime import PromptRuntimeComponentsBuilder from a2a_t.config.models import A2ATConfig from a2a_t.prompt.analysis import ScenarioRecognizer, ScenarioResolutionOrchestrator, SlotExtractor from a2a_t.prompt.task_rendering import TaskPr...
project-openan/a2a-t-sdk-python
src/a2a_t/client/prompt_generation/prompt_generation_orchestrator_builder.py
.py
36c7a8e2dfad7ed0
7.66
20
from __future__ import annotations class PromptResourceError(Exception): """Base class for shared prompt resource loading errors.""" def __init__(self, message: str, **context: object) -> None: super().__init__(message) self.context = context class PromptResourceNotFoundError(PromptResource...
project-openan/a2a-t-sdk-python
src/a2a_t/common/prompt_resources/errors.py
.py
dd42f0a9ce5fdb2d
7.66
20
from __future__ import annotations import json from pathlib import Path from typing import Any from a2a_t.common.resource_roots import resolve_prompt_resource_root from a2a_t.prompt.common.errors import PromptSourceError from .errors import PromptResourceNotFoundError, PromptResourceParseError def _packaged_prompt...
project-openan/a2a-t-sdk-python
src/a2a_t/common/prompt_resources/local_resources.py
.py
c9ead428c4d68eb7
7.66
20
from __future__ import annotations from .local_resources import BasePromptResourceLoader from .models import PromptMessages class PromptResourceLoader(BasePromptResourceLoader): """Load shared analysis prompts used by LLM-powered prompt analysis.""" def load(self, *, analysis_action: str, language: str) -> ...
project-openan/a2a-t-sdk-python
src/a2a_t/common/prompt_resources/prompt_resource_loader.py
.py
ffe1ab7f33aa70b5
7.66
20
from __future__ import annotations from .local_resources import BasePromptResourceLoader from .models import ScenarioDefinition class ScenarioLoader(BasePromptResourceLoader): """Load scenario definitions for scenario recognition.""" def load(self, *, language: str) -> list[ScenarioDefinition]: """R...
project-openan/a2a-t-sdk-python
src/a2a_t/common/prompt_resources/scenario_loader.py
.py
077a015edd30fd6c
7.66
20
from __future__ import annotations from a2a_t.prompt.common.models import PromptReference from .errors import PromptResourceParseError from .local_resources import BasePromptResourceLoader from .models import SlotDefinition, SlotRange, SlotSchema class SlotSchemaLoader(BasePromptResourceLoader): """Load slot sc...
project-openan/a2a-t-sdk-python
src/a2a_t/common/prompt_resources/slot_schema_loader.py
.py
dda4d20fb3e2aacf
7.66
20
from __future__ import annotations from a2a_t.prompt.common.models import PromptReference from .local_resources import BasePromptResourceLoader class TemplateLoader(BasePromptResourceLoader): """Load scenario-specific task prompt templates.""" def load(self, *, reference: PromptReference) -> str: "...
project-openan/a2a-t-sdk-python
src/a2a_t/common/prompt_resources/template_loader.py
.py
2ef9b442537309e0
7.66
20
"""Let a repair round say *what to change* instead of retyping the file. The developer's output contract had one verb: here is a file, here are its full contents. So every repair regenerated whole files, and a regenerated file is retyped from the model's memory of it — which is where invented names come from. Straight...
alexar76/aicom
agents/dev_edits.py
.py
74560f02d1a5bb53
7.42
6
""" Product delivery profile — drives PM depth, spec gate, and architect emphasis. """ from __future__ import annotations import os import re from core.delivery_profile import DESKTOP_APP, FULL_SOFTWARE, MARKETING_LANDING, normalize_delivery_profile __all__ = [ "DESKTOP_APP", "FULL_SOFTWARE", "MARKETING_...
alexar76/aicom
agents/product_profile.py
.py
6d770beaf80f51ec
7.42
6
"""The emitter must produce documents the reference verifier accepts, and nothing else. Every assertion here goes through `awr.verify_document` rather than inspecting fields by hand: the emitter's contract is "the result verifies", not "the dict looks right". """ from __future__ import annotations import json import...
alexar76/aicom
awr/emitters/python/tests/test_emitter.py
.py
9130e008a867b1cb
7.92
6
"""SHA-256 digests and W3C SRI digest references (SPEC.md sections 3.2, 8.1). An SRI string in AWR/2 is exactly ``sha256-`` followed by *standard, padded* base64 (``+/`` alphabet) of the 32-byte digest. base64url is not accepted: a digest reference is inside the signature, so accepting a second spelling of the same v...
alexar76/aicom
awr/reference/python/awr/digest.py
.py
f13f5ba842f57403
7.42
6
"""Reason-code registry and the single exception type of the AWR reference implementation. The registry is a transcription of AWR/2 SPEC.md section 11.2. Codes are stable: a code is never re-used for a different meaning (section 11.1). """ from __future__ import annotations from typing import Dict, List, NamedTuple...
alexar76/aicom
awr/reference/python/awr/errors.py
.py
dedb0f7d3dd17308
7.42
6
"""RFC 8785 JSON Canonicalization Scheme, profiled by AWR/2 SPEC.md section 4. Two things in here are the reason this module is hand-written instead of delegating to a library: 1. Property names are sorted as **arrays of UTF-16 code units compared as unsigned integers** (RFC 8785 section 3.2.3). Python's ``sorted...
alexar76/aicom
awr/reference/python/awr/jcs.py
.py
eda0097c2eb7318b
7.42
6
"""AWR/1 legacy verification (SPEC.md section 12). AWR/1 documents carry ``Ed25519Signature2018``, a base64 ``proofValue``, and a signature over a pipe-delimited rendering of ``credentialSubject`` **only**. Consequences that this module enforces: * ``id``, ``type``, ``issuer`` and any ``hubInfo`` are **unsigned** an...
alexar76/aicom
awr/reference/python/awr/legacy.py
.py
1e06ebc5d8c7bf85
7.42
6
"""Hand-written base58btc and the multibase ``z`` prefix (SPEC.md section 5.1, 6.1). base58btc is needed in two places -- the ``did:key`` method-specific identifier and ``proof.proofValue`` -- and is implemented here rather than pulled in as a dependency so that the reference implementation has no dependency beyond ``...
alexar76/aicom
awr/reference/python/awr/multibase.py
.py
c95fe2f4cfc4dfd5
7.42
6
"""ABO's seeded simulation: the floor, the four styles, and what it measures. These run tiny samples. The 1,000-completed-game floor is a property of the gate, not of the sample size a check can afford, so it is proved by handing the gate a sample that falls one short rather than by playing a thousand games in a unit ...
autonomous-ai/autonomous-workshop
inventors/abo/skills/abo-inventor/scripts/tests/test_simulation.py
.py
893586846a33b1fc
8.04
11
"""Opt-in Git push for one generated toy directory.""" from __future__ import annotations import os import subprocess from pathlib import Path from typing import Mapping, Optional from workshop.errors import EffectError, StateConflict class GitPushError(EffectError): """The requested Git command failed.""" d...
autonomous-ai/autonomous-workshop
src/workshop/integrations/git.py
.py
2fe7df1ab1986e6e
7.54
11
#!/usr/bin/env python """FDM mating fits — one source of truth for assembled-part clearances. A mating interface (tab/slot, lip/groove, peg/socket, pin/hole, lid/cavity) has a male and a female half that **must** derive from one nominal dimension plus a clearance, or they drift until the parts jam or fall apart. Size ...
autonomous-ai/autonomous-workshop
src/workshop/make/skills/cad/scripts/cadfits.py
.py
f69cb9f34a6c7871
7.54
11
"""Stdlib-only client for the warm CAD CLI daemon. The tool launchers' ``CADGEN_WARM`` shim imports this module BEFORE any heavy import, so it must stay dependency-free and cheap to import. Everything here falls back to ``None`` (caller runs inline, cold) on any spawn or protocol problem — the daemon is a fast path, n...
autonomous-ai/autonomous-workshop
src/workshop/make/skills/cad/scripts/cadgen_daemon/client.py
.py
f12f40c4d0b744f1
7.54
11
"""Warm-process daemon server for the CAD skill CLIs. One long-lived process imports cadgen / OCP / build123d ONCE and then services ``scripts/gen`` / ``scripts/export`` / ``scripts/artifact`` / ``scripts/inspect`` invocations over a socket shared by one materialized CAD skill tree, so opted-in sessions (``CADGEN_WARM...
autonomous-ai/autonomous-workshop
src/workshop/make/skills/cad/scripts/cadgen_daemon/server.py
.py
3f6fda5c75eebb17
7.54
11
#!/usr/bin/env python3 """Print-parameter derivations for FDM, and the hollow that actually hollows. The sibling of `cadfits`. That one derives the second half of a mate from the first so a clearance cannot be typed twice; this one derives a wall from the nozzle that has to lay it down, and removes material the same w...
autonomous-ai/autonomous-workshop
src/workshop/make/skills/cad/scripts/cadprint.py
.py
15123dad21e7948d
7.54
11
""" Shared machinery of the two chDB (embedded ClickHouse) in-memory executors. Both run the same ClickHouse SQL from artifacts/queries/inmemory/chdb.psv in one chDB session. They differ in where the data they query lives -- which the query file's {name} table placeholders abstract away -- and, with it, in the form a ...
KxSystems/NYSETAQBenchmarks
pysrc/queryrunner/executors/inmemory/chdb_base.py
.py
1aedb414216aada2
7.5
9
""" chDB (embedded ClickHouse) in-memory query executor over Apache Arrow tables. The tables (master, trade, quote, exnames, timeBuckets) are stored in memory as Apache Arrow tables. The queries reference them through chDB's Python() table function (e.g. SELECT ... FROM Python(trade)), which scans the Arrow data direc...
KxSystems/NYSETAQBenchmarks
pysrc/queryrunner/executors/inmemory/chdb_pyarrow.py
.py
eae4d771bda2a818
7.5
9
""" Polars in-memory executor running the eager API (single partition loaded into memory): the tables are DataFrames, every setup step runs eagerly, and a query returns a DataFrame. The shared machinery is in polars_base.py. """ from pathlib import Path import polars as pl from executors.inmemory.polars_base import Q...
KxSystems/NYSETAQBenchmarks
pysrc/queryrunner/executors/inmemory/polars_eager.py
.py
f792f93d367401ca
7.5
9
""" Polars in-memory executor running the lazy API with the streaming engine (single partition loaded into memory): the queries see LazyFrames and their results are collected with the streaming engine, as are the load and transform steps. The sort stays on the in-memory engine, which is faster for it. The shared machin...
KxSystems/NYSETAQBenchmarks
pysrc/queryrunner/executors/inmemory/polars_lazy_streaming.py
.py
a04f7be9393e7c06
7.5
9
import logging from datetime import datetime from typing import List, Dict import pyarrow as pa import pyarrow.compute as pc def letter_filter(start_char: str, end_char: str, table: pa.Table) -> pa.Table: """Filters a table based on the first letter of the 'Symbol' column. Includes rows where the first chara...
KxSystems/NYSETAQBenchmarks
pysrc/taqToParquet/table_converters.py
.py
6791b3b3fbcf68d0
7.5
9
#!/usr/bin/env python3 """Import every shipped module on the OLDEST Python this repo supports. `pyproject.toml` declares `requires-python = ">=3.9"` and the README promises "python3 (3.9 or newer)". These skills are publicly installed and run against whatever `python3` a user happens to have — while every workflow her...
starslingdev/skills
.github/scripts/check_python_floor.py
.py
31d18019c93d41ce
7.65
19
"""One place that knows what the scanner says, so a rename breaks one file loudly. The workflow reads the scanner's answer in three independent places — the gate decides pass/fail, the reporter surfaces warnings, the red-proof control checks the scanner can still see. Each used to hardcode its own understanding of the...
starslingdev/skills
.github/scripts/registry_scan_contract.py
.py
acdb16c1a6f1d3ea
7.65
19
#!/usr/bin/env python3 """Prove the registry-scan gate can actually go red. A check that cannot fail is not a check. `.github/workflows/registry-scan.yml` scans the skills under `skills/` and fails the build on a finding — but if the scanner stopped detecting the class of problem we care about, the gate would go on pa...
starslingdev/skills
.github/scripts/registry_scan_redprove.py
.py
b4feb28bd9977f1f
7.65
19
#!/usr/bin/env python3 """Surface every registry-scan finding, without letting a warning block the build. `.github/workflows/registry-scan.yml` blocks on every risk 0.6.0 reports except the ones named in `registry_scan_contract.NON_BLOCKING_RISKS`. Those exempt risks are real findings we want a human or a review agent...
starslingdev/skills
.github/scripts/registry_scan_report.py
.py
14122dda36631ca5
7.65
19
"""A session that never started must not be graded as a session that misbehaved. The first CI run of the eval workflow failed every grader on all five cases in 1.6 seconds with zero tool calls: `claude` refused to start under `--permission-mode bypassPermissions` as root and exited immediately, printing its reason on ...
starslingdev/skills
maintainers/ci-secure/tests/test_run_skill_evals.py
.py
74ee4228cd4d4c53
7.15
19
#!/usr/bin/env python3 """Phase 4c helper — turn captured catalog gaps into drafted detectors. SKILL.md phase 4b captures every coverage-gap pole (a drilled job whose log matched no `_parse_log` detector) to `.ci-speedup-gaps/<repo>__<job>/`. Phase 4c (maintainer source only) promotes those one-off LLM gap-fills into ...
starslingdev/skills
maintainers/ci-speedup/scripts/draft_detector.py
.py
d8180a6b494f7961
7.65
19
#!/usr/bin/env python3 """(Re)install the /ci-speedup-dogfood slash command from its canonical body. The committed canonical body lives at `maintainers/ci-speedup/workflows/ci-speedup-dogfood.command.md`; the *installed* command Claude Code actually runs lives at the gitignored `.claude/commands/ci-speedup-dogfood.md`...
starslingdev/skills
maintainers/ci-speedup/scripts/install_dogfood_command.py
.py
728b90b11adedae6
7.65
19
#!/usr/bin/env python3 """Phase-0 baseline producer (maintainers-only) — seal-single-door.md §4(B). Runs the cross-seam **contradiction property** (`verify_report`'s Phase-0 checks) over a panel of already-rendered reports and prints the two baseline rates that the relocation **Decision gate** (§4) consumes: - **co...
starslingdev/skills
maintainers/ci-speedup/scripts/measure_contradictions.py
.py
05db51a3be4aae64
7.65
19
"""Pin the /ci-speedup-dogfood slash command's entry point to a workflow that exists. The command is installed by copying the canonical body (`maintainers/ci-speedup/workflows/ci-speedup-dogfood.command.md`) into the gitignored `.claude/commands/` dir. That body launches the loop via `workflow({ scriptPath: '<path>' }...
starslingdev/skills
maintainers/ci-speedup/tests/test_dogfood_command_integrity.py
.py
7b5109ed1228228e
8.15
19
"""Tests for `draft_detector.py` — the phase-4c gap → catalog helper. The script's value is that its three deterministic halves can't drift from what the renderer does: a capture is "pending" iff `_parse_log` returns None on its real job log (same ground truth), `prepare` emits a task only for pending gaps, and `verif...
starslingdev/skills
maintainers/ci-speedup/tests/test_draft_detector.py
.py
211d4146373ef4c0
8.15
19
"""Pin the /ci-speedup-dogfood installer's transform so a refresh can't corrupt the command. `install_dogfood_command.render_installed_body` turns the committed canonical body into the form written to the gitignored `.claude/commands/`. These tests exercise that transform purely (no `.claude/` needed, so CI runs them)...
starslingdev/skills
maintainers/ci-speedup/tests/test_install_dogfood_command.py
.py
55c3516138ed9249
8.15
19
"""ci-score entry point (B1) — collect a repo's CI configuration and stamp the CI Score, offline, from a local checkout only. Flow: verify the target is a git checkout → read + parse `.github/workflows/*.yml|yaml` → `_practice_facts(parsed, root)` (which itself walks local composite actions and the repo-root build-too...
starslingdev/skills
skills/ci-score/scripts/collect_config.py
.py
a53dcbcf689a193b
7.65
19
"""ci-score report invariant checker (B2) — one command, PASS/FAIL. `verify(doc, report_text, registry)` asserts the report faithfully renders its findings document. Verifying a report must never be manual archaeology: CI, the dogfood sweep, and a human all run the same checks and get the same verdict. Invariants (eac...
starslingdev/skills
skills/ci-score/scripts/verify_report.py
.py
21f97739304dc355
7.65
19
"""Materialize the cloaked third-party control fixtures back to their exact original trees at test time. The two frozen control checkouts (mastra, better-auth) are real third-party repo snapshots — workflow YAML, ``.github/actions/**`` composite actions, and root build-tool configs, exactly the input surface the score...
starslingdev/skills
skills/ci-score/tests/_fixture_checkouts.py
.py
42a9662cf86f263b
8.15
19
"""B4 layer (a): exact-match controls on frozen fixture checkouts. The calibration positive controls (mastra 82 B+, better-auth 82 B+) must reproduce EXACTLY from the committed fixture trees under `tests/fixtures/checkouts-src/` — the full collector path (acquire → facts → score), not just the arithmetic. Live repos d...
starslingdev/skills
skills/ci-score/tests/test_b4_controls.py
.py
bfb4493015300295
8.15
19
"""The relocated close-honesty pin (moved from ci-speedup's test_close_guidance by the score-ectomy, 2026-07-16). ci-score's close opens with the CI Score read straight off the stamp and NEVER fabricates a score — the no-score cases (refusal / `ci_score_error` / absent stamp) are each handled honestly. The close is LL...
starslingdev/skills
skills/ci-score/tests/test_close_contract.py
.py
1fbedeb998b9e333
8.15
19
"""Byte-fidelity + cloak-property proof for the frozen control fixtures. The two third-party control checkouts (mastra, better-auth) ship "cloaked" so registry security scanners don't attribute their workflow files to this repo: under ``tests/fixtures/checkouts-src/<name>/`` every ``.github`` path segment is renamed t...
starslingdev/skills
skills/ci-score/tests/test_fixture_cloak.py
.py
c9ec53aca99f55e0
8.15
19
"""Guards for the OD-L4 prompt-routing contract (B3). The routing itself is LLM-driven (skill descriptions), so the committed artifact + the SKILL.md description are the only guardable surfaces. These cells pin: the artifact exists and is well-shaped; OD-L4's required routings hold (ambiguous middle → ci-speedup; expl...
starslingdev/skills
skills/ci-score/tests/test_routing_artifact.py
.py
56c8277a0a53a975
8.15
19
#!/usr/bin/env python3 """ Bounded Concurrency Demonstration This script sends a bounded batch of synthetic evidence-triage requests to the Cloudflare Worker. It reports observed task states and latency for this run. It does not establish capacity, cost savings, production readiness, or equivalence to human review. U...
vassiliylakhonin/agenda-intelligence-md
examples/infinite-swarm-batch.py
.py
3a7507bb4fb4bcdf
7.42
6
"""DFIR JSON line output module for Plaso. A drop-in psort output module (`-o l2t_json_dfir`) that emits the same rich, per-parser json_line as the built-in `json_line`, but adds four attribution fields resolved FROM THE .plaso DB — so the output is self-describing and no second pass over the logs is needed: image_...
Get-Sybers/DX_DFIR
dev-scripts/plaso/l2t_json_dfir.py
.py
ef88838f331a082b
7.45
7
"""Volatility 3 output renderer: flat JSON Lines for the DFIR pipeline. `-r jsonl_dfir` — emits ONE JSON object per TreeGrid node, not per tree-root the way the built-in `JSONL` (JsonLinesRenderer) does. The built-in keeps the tree nesting (`__children`), so a `pstree` line is a whole subtree; this flattens it, so eve...
Get-Sybers/DX_DFIR
dev-scripts/volatility/jsonl_dfir_renderer.py
.py
565ab444c3151836
7.45
7
"""Volatility 3 plugin: DFIR process records for the CAR data model. `dfir_processes.DfirProcesses` — one row per process, enumerated with psscan (pool-tag scanning) so UNLINKED / terminated processes a rootkit hid from the active list are still found. Each row carries what CarProcess wants and no single built-in plug...
Get-Sybers/DX_DFIR
dev-scripts/volatility/plugins/dfir_processes.py
.py
2359fe3698588f3a
7.45
7
"""Targeted artefact extraction from disk images, via log2timeline/plaso. The evtx lane needs real ``.evtx`` files, but evidence often arrives as a disk image (E01/raw/VMDK), not loose logs. This module pulls ONLY a named artefact set out of an image with Plaso's ``image_export.py`` (dfVFS, userspace, E01-capable) — t...
Get-Sybers/DX_DFIR
python/get_sybers_dfir/imageexport.py
.py
bfba164398846f98
7.45
7
"""Tool-image inventory guard — refuse to run against anything but the known, hardened dfir/* images, and flag anything added to that namespace that should not be there. Two checks: - ``require(image)`` — run at the START of every processor (its role preflight): the exact image the lane is about to run must exist, ...
Get-Sybers/DX_DFIR
python/get_sybers_dfir/images.py
.py
e12b776116889609
7.45
7
"""Record-shaping for ingest — the constant-column wrapping ``.ingest`` cannot do. ``.ingest into`` cannot inject a per-file constant (which log type / plugin / artefact / image a row came from), and it cannot convert Plaso's microsecond epoch to a datetime. So each source's files are rewritten to JSON Lines with the ...
Get-Sybers/DX_DFIR
python/get_sybers_dfir/ingest/prepare.py
.py
fd368a87f0cc93fe
7.45
7
"""Signature/detection processors — YARA, Suricata, Hayabusa. Port of the retired ``process-signatures.sh`` (+ its ``signatures/{yara,suricata, hayabusa}.sh`` lanes and ``lib/disk-image.sh``). Runs the three signature engines over the evidence and lands their native events as ingest-ready JSON Lines under ``<output_di...
Get-Sybers/DX_DFIR
python/get_sybers_dfir/signatures/__init__.py
.py
617ae3a78c8d8201
7.45
7
"""DetectRaptor YARA provisioning — fetch, verify, merge into one ruleset. `mgreen27/DetectRaptor <https://github.com/mgreen27/DetectRaptor>`_ ships bulk Velociraptor detection content. Most of it (VQL artifacts + the CSV lookups that drive them) needs a Velociraptor server and has no consuming lane here; what THIS pi...
Get-Sybers/DX_DFIR
python/get_sybers_dfir/signatures/detectraptor.py
.py
1683ccfee66ce61d
7.45
7
"""Hayabusa lane — Sigma-based Windows event-log detection -> native JSONL. Scans loose ``*.evtx`` under the raw tree AND disk images, standalone — no fuse mount and no prior processor run required. Disk-image EVTX come from the SHARED staged extraction (``imageexport.extract_staged``: log2timeline/plaso ``image_expor...
Get-Sybers/DX_DFIR
python/get_sybers_dfir/signatures/hayabusa.py
.py
0635ba56c41f9d82
7.45
7
"""Velociraptor processor — lay out offline-collector ZIPs for ingestion. Unpack each collection ZIP and lay its per-artefact ``results/<Artefact>.json`` (JSON Lines) out flat under one folder per collection, where the ingest loader (``get_sybers_dfir.ingest``) expects it. The ingest-side wrap does the ``{Artefact, So...
Get-Sybers/DX_DFIR
python/get_sybers_dfir/velociraptor.py
.py
35fd1f3d3fc55afb
7.45
7
"""Volatility 3 processor — memory images -> per-plugin JSON Lines. Run a fixed set of Volatility 3 plugins over each memory image with the custom ``jsonl_dfir`` renderer (one flat JSON object per TreeGrid node — one process/connection/artefact per line), writing ``<plugin>.jsonl`` per image. The ingest loader wraps e...
Get-Sybers/DX_DFIR
python/get_sybers_dfir/volatility.py
.py
0d4dc07ac500d359
7.45
7
"""Zeek processor — PCAPs → Zeek JSON logs. Discover captures anywhere under a pcap tree (magic bytes first, extension as fallback), run the ``zeek/zeek`` container per capture with ``LogAscii::use_json=T`` + ISO-8601 timestamps, and rename each ``*.log`` (Zeek keeps the extension even for JSON) to ``*.json`` under on...
Get-Sybers/DX_DFIR
python/get_sybers_dfir/zeek.py
.py
773820b4a66cb423
7.45
7
"""Unit tests for the DetectRaptor provisioning module (pure logic, no network).""" import os from get_sybers_dfir.signatures import detectraptor _A = ( 'import "pe"\n' "rule Alpha : FILE {\n" " strings:\n $s = \"x\"\n condition:\n $s\n}\n" "private rule Helper {\n condition:\n...
Get-Sybers/DX_DFIR
python/tests/test_detectraptor.py
.py
5f29d2408acef74a
7.95
7
"""Unit tests for the SOF-ELK delivery module (local copy + ledger idempotence).""" import os from get_sybers_dfir import sofelk def _seed(src): os.makedirs(os.path.join(src, "zeek", "cap")) with open(os.path.join(src, "zeek", "cap", "conn.json"), "w") as fh: fh.write('{"id":1}\n') with open(os.p...
Get-Sybers/DX_DFIR
python/tests/test_sofelk.py
.py
f925d96e7f5f6450
7.95
7
"""Unit tests for the velociraptor processor (pure logic + real zipfile, no tool).""" import json import os import zipfile from get_sybers_dfir import velociraptor def _make_zip(path, entries): with zipfile.ZipFile(path, "w") as zf: for name, data in entries.items(): zf.writestr(name, data) ...
Get-Sybers/DX_DFIR
python/tests/test_velociraptor.py
.py
7601b12a4963f475
7.95
7
"""Glyph id lookup for tests that build a :class:`torchfont.GlyphRef` by hand. ``GlyphRef`` names a glyph by id. ``CodepointDataset`` resolves ids from codepoints while indexing, so tests that skip the dataset resolve them here with fontTools instead. """ from __future__ import annotations from fontTools.ttLib impor...
torchfont/torchfont
tests/_glyphs.py
.py
b08bf1bbd4200f91
8.13
17
"""Tensor-pair adapters for kernels whose public API takes an ``Outline``. Many kernel tests were written against ``(types, coords)`` pairs. The public functional API takes and returns :class:`torchfont.Outline`, so these adapters keep those tests focused on kernel behaviour instead of container plumbing. """ from __...
torchfont/torchfont
tests/_pairs.py
.py
09d8fee673ced04c
8.13
17
import logging from pathlib import Path import pytest import torch from torch import Tensor from torch.utils.data import DataLoader from tests._pairs import remove_overlaps from torchfont import CodepointSample, ElementType, Outline from torchfont.datasets import CodepointDataset from torchfont.transforms import func...
torchfont/torchfont
tests/google_fonts/test_remove_overlaps.py
.py
a47c810915835db4
7.13
17
"""Unit tests for the force-push classification and the guard's target parsing. The default-branch guard now runs before the HTTPS push as well, which means it has to recognise a history rewrite from the argument list alone. These tests pin that down, plus the refspec parsing that decides *which* remote branch the pus...
phantomyard/phantomtools
github-app-auth/tests/test_force_args.py
.py
665e85853273e219
7.92
6
"""SQLite FTS5 full-text index for an OKF bundle (H1). Replaces okf.search()'s str.count() scoring with a proper inverted index. Design: - One index.db per agent, at <agent_dir>/.fts/index.db (inside the bundle, git-trackable, not a concept dir so it never shows up as knowledge). - Tokenizer: unicode61 with remove_d...
sillogic/sill-ensoul
ensoul/fts.py
.py
a35974001586ab81
7.42
6
"""sill-ensoul HTTP MCP server: the same 8 OKF tools over Streamable HTTP, protected by Bearer-token auth (SIL-7). Purpose: deploy the knowledge base as a REMOTE MCP server (cloud VPS / tailnet) so multiple machines share ONE memory. Authentication is the connection layer — "who may connect". Every request must carry ...
sillogic/sill-ensoul
ensoul/http.py
.py
39b55c5d453fcd49
7.42
6
"""`sill-ensoul-init` console script: initialize an empty global KB and manage per-CLI shell integration. Designed for sharing sill-ensoul with teammates: they `pip install` the package, run `sill-ensoul-init`, then follow the printed steps to wire it into their CLI. The KB starts empty — each user builds their own ag...
sillogic/sill-ensoul
ensoul/init_cmd.py
.py
e6a9ddccb1ce29a7
7.42
6
"""sill-ensoul MCP server: exposes an OKF knowledge base as agent tools. Run as a stdio MCP server: python -m ensoul.server Register it from Codex / Claude Code / OpenCode (see README.md). Terminology: an "agent" in the tool names below is what users call an "ensouler" (a role-scoped memory bundle). The MCP tool nam...
sillogic/sill-ensoul
ensoul/server.py
.py
2b7dcefa09f4b38f
7.42
6
"""One-command test runner for sill-ensoul. Usage: python -m tests.run_tests # from repo root (preferred) python tests/run_tests.py # also works """ import subprocess import sys from pathlib import Path # Repo root = parent of this tests/ directory. Tests are imported as # `tests.test_*`, so cwd mus...
sillogic/sill-ensoul
tests/run_tests.py
.py
1608f8d367240f69
7.92
6
"""搜索回归测试:防 H1(FTS5 检索)/H7(persona 排除)退化。 自建隔离的临时 agent,写多条 concept,验证检索行为,跑完清理。 不依赖真实 algo-engineer 数据,可任意次数重复运行。 用法: python -m tests.test_search """ import os import shutil import tempfile import yaml from ensoul import okf, fts AGENT = "search-regression-fixture" # 临时 agent,跑完删除 def setup_agent(kb_root): ...
sillogic/sill-ensoul
tests/test_search.py
.py
eb623606558d4055
7.92
6
"""Circuit breaker - fail-fast protection for a slow/failing LLM provider. State is process-global and keyed by ``(provider, model)`` so each candidate model accumulates its own failure history independently. Transitions follow the standard closed -> open -> half-open cycle: - **CLOSED**: calls allowed. Each fallback...
PrisonBreakPB/Folium
folium/circuit_breaker.py
.py
71d1fe58ff1c6ce0
7.45
7
"""Configuration - env vars and defaults.""" import os import re from dataclasses import dataclass from pathlib import Path DEFAULT_MAX_CONTEXT_TOKENS = 1_000_000 @dataclass class LLMProfile: """A named, self-contained LLM endpoint configuration.""" name: str provider: str api_key: str base_ur...
PrisonBreakPB/Folium
folium/config.py
.py
c824e26ebece127a
7.45
7
"""Cost meter - per-session rolling budget for LLM spend. Owned by the agent that drives a task/session; every LLM call (main loop, context compression, background memory maintenance) reports its cost here so the agent can watch how much it has spent. Budgets are in dollars. A ``budget_usd`` of 0 disables the meter (s...
PrisonBreakPB/Folium
folium/cost_meter.py
.py
57794aa723206789
7.45
7
"""LLM gateway - control plane for model selection (rule-based routing). Holds everything that decides *which* model to call: scene -> tier candidates and tier -> concrete model id. Transport (retry/streaming/errors) stays in ``llm.py``; callers only report their ``scene`` and do not pick a model. """ from __future__...
PrisonBreakPB/Folium
folium/gateway.py
.py
428d8203ad11c948
7.45
7
"""Trace event recorders.""" from __future__ import annotations import json import threading from pathlib import Path from typing import Any from ..database import get_connection class NullRecorder: def record(self, event: dict[str, Any]) -> None: return None def flush(self, trace_id: str) -> None...
PrisonBreakPB/Folium
folium/observability/recorder.py
.py
7f392391058997a4
7.45
7
"""SQLite trace query helpers.""" from __future__ import annotations import json from pathlib import Path from ..database import get_connection def _db_path(value: str | Path | None) -> Path | None: if value is None: return None path = Path(value) return path if path.suffix == ".db" else path /...
PrisonBreakPB/Folium
folium/observability/summary.py
.py
63a5a8c7083989cb
7.45
7
"""System prompt - the instructions that turn an LLM into a research assistant.""" import os import platform from pathlib import Path MEMORY_FILE = Path(r"D:\learn-Agent\Folium\memory.md") MAX_MEMORY_CHARS = 2000 _ROLE_FALLBACK = """\ You are Folium, an AI research assistant designed to support the full academic res...
PrisonBreakPB/Folium
folium/prompt.py
.py
be39f6c9f5f457f1
7.45
7
"""Filesystem routing for local and sandbox-backed tools.""" from pathlib import Path from .session import SandboxPathError, get_current_session, get_host_workspace, use_copy_workspace def resolve_tool_path(file_path: str) -> Path: if use_copy_workspace(): return get_current_session().resolve_path(file_p...
PrisonBreakPB/Folium
folium/sandbox/filesystem.py
.py
927b1f8e3e1b825a
7.45
7
"""SQLite-backed session persistence.""" from __future__ import annotations import copy import json import re import time import uuid from pathlib import Path from typing import Any from .database import get_connection from .encoding import repair_mojibake_text _SAFE_SESSION_RE = re.compile(r"[^A-Za-z0-9._-]+") _SU...
PrisonBreakPB/Folium
folium/session.py
.py
dca5261b55bafa92
7.45
7
"""Base class for all tools.""" from abc import ABC, abstractmethod from dataclasses import dataclass from typing import ClassVar from pydantic import BaseModel, ValidationError @dataclass class ToolValidationError(Exception): """Raised when a tool call does not match the tool's JSON schema.""" tool_name: ...
PrisonBreakPB/Folium
folium/tools/base.py
.py
015bf42f4b8f5832
7.45
7
"""Content search with regex support.""" import re from pathlib import Path from pydantic import BaseModel, ConfigDict, Field from .base import Tool, ToolFailure, tool_failure from ..sandbox.filesystem import SandboxPathError, resolve_tool_path # skip these dirs to avoid noise _SKIP_DIRS = {".git", "node_modules", ...
PrisonBreakPB/Folium
folium/tools/grep.py
.py
02ff9fed9d95dcf8
7.45
7
# Copyright 2026 The HuggingFace Team. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicabl...
huggingface/transformers-mlinter
mlinter/_helpers.py
.py
a8618a5375170391
7.42
6
# Copyright 2026 The HuggingFace Team. All rights reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicabl...
huggingface/transformers-mlinter
mlinter/trf003.py
.py
0591489e10228d1a
7.42
6