markitect-main/markitect/infospace
tegwick 1b9a31665c fix(pipeline): retry on all LLM errors (not just rate limits)
Free-tier APIs intermittently return invalid JSON or empty responses.
Now any exception in _call_llm retries up to 3 times with a 5s back-off,
rather than failing immediately on non-rate-limit errors.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-02-19 20:32:23 +01:00
..
checks feat(infospace): add collection-level quality checks C1–C5 (S2.4) 2026-02-19 01:54:22 +01:00
__init__.py feat(infospace): add infospace configuration model and state (S2.1) 2026-02-19 01:44:14 +01:00
cli.py feat(pipeline): per-stage max_tokens, LLM provenance, processing log 2026-02-19 14:50:49 +01:00
composition.py feat(infospace): add composition model for discipline binding (S2.6) 2026-02-19 02:03:54 +01:00
config.py feat(pipeline): per-stage max_tokens, LLM provenance, processing log 2026-02-19 14:50:49 +01:00
entity_parser.py feat(infospace): add entity metadata parser (S1.1) 2026-02-19 00:27:45 +01:00
evaluate.py feat(infospace): add per-entity evaluation pipeline and CLI command (S2.3) 2026-02-19 01:48:34 +01:00
evaluation.py feat(infospace): add structured evaluation output with history and diffing (S1.5) 2026-02-19 01:35:22 +01:00
evaluation_io.py feat(infospace): add structured evaluation output with history and diffing (S1.5) 2026-02-19 01:35:22 +01:00
history.py feat(infospace): add metrics history and viability tracking (S2.5) 2026-02-19 02:01:00 +01:00
models.py feat(infospace): add entity metadata parser (S1.1) 2026-02-19 00:27:45 +01:00
pipeline.py fix(pipeline): retry on all LLM errors (not just rate limits) 2026-02-19 20:32:23 +01:00
schema.py feat(infospace): add schema compliance validator (S1.2) 2026-02-19 00:48:57 +01:00
state.py feat(infospace): add infospace configuration model and state (S2.1) 2026-02-19 01:44:14 +01:00
validator.py feat(infospace): add schema compliance validator (S1.2) 2026-02-19 00:48:57 +01:00