Spaces:
Running
Running
Commit History
16 days ago
-
Convert newlines to br in rich path so lists stop collapsing
SilverElixiraa3af88 -
Presplit glued bullets before chunking so lists survive cutting
SilverElixir1dfad36 -
Split short platform bullet lines to fix mashed lists
SilverElixird926362
17 days ago
-
Fix adversarial review findings across splitter quotas and backup
SilverElixir2939738 -
Split messages by HTML length in UTF-16 units
SilverElixir1190087 -
Bound table separator match length for pathological lines
SilverElixire89951a
18 days ago
-
Merge near-duplicate formatting tests into parametrized cases
SilverElixir87f3899 -
Drop 4 duplicate and introspection-only tests
SilverElixirfd5cff9 -
Split glued bullet lists without spaces or with middot
SilverElixir281f95c
25 days ago
-
Subagent re-review: fix confirmed misses, harden thin tests
SilverElixirFour parallel adversarial reviewers re-checked the wave. Confirmed and fixed: - lumen_routes: a failed MALFORMED_FUNCTION_CALL retry returned the "response blocked" placeholder as a successful answer and stopped the chain. Retry failure now falls through to the next model like the spent-budget path. - lumen_message_core/bot.py: lock-busy replies and pick language fallback no longer create chat state via get_state (_t_no_create/_peek_chat_lang). - lumen_commands: a pick tap without a message no longer burns the token. - bot.py: lock comment now states the truth — slash commands bypass the lock, draw/tts trigger phrases run under it. - lumen_formatting: table-block lines without a leading pipe are skipped by both inline splitters, closing the HTML/rich divergence on bare tables. Tests hardened (each fails without its fix where stated): timeout fallback now uses a real async hang with distinct answers; budget assertion tightened; deadline test bounds elapsed time; admin seed f…3d3384a
26 days ago
-
Fix regressions found by adversarial review: routing, rendering, rate limits
SilverElixirRegressions from the previous waves, found by six independent adversarial reviews and confirmed directly: - lumen_routes: the account-wide OpenRouter early exit had moved into the request wrapper, where the shared except swallowed it, so the whole chain was burned instead of stopping after one attempt. The decision now lives in the loop via stop_on. - lumen_routes: a skipped retry after a malformed tool call used to answer the user with a "response blocked" placeholder instead of falling through to the next model. - lumen_formatting: the inline list splitters ate blockquotes, headings and markdown table rows, so quotes rendered empty, a bare heading marker leaked and tables broke differently in the stream and in the final message. - lumen_formatting: the element-of command preceded the integral command in the symbol map, so int with limits turned into element-of plus t. Headings were also extracted before the LaTeX scrub, leaving raw commands in h3. - lumen_streaming…62c1832 -
Quality wave 5: summarizing history trim in streaming, LaTeX parity between render paths, rate limit before state/album downloads, parity tests
SilverElixir5a68db4
27 days ago
-
Split glued 3-item bullet lists in both render paths (prod 26.09.2026 threshold miss)
SilverElixir76aeacc
28 days ago
-
Claude-parity fixes: callback filters for pick buttons, inline numbered-list splitter in both render paths, prompt ban on glued lists
SilverElixird59ebea
30 days ago
-
Formatting like Claude: one list item per line, prompt rule plus post-fix
SilverElixir693519a
-
Move tests to tests/ and proxy to proxy/, update CI and docs paths
SilverElixir7a11b13