Celes Renata 1394e6168b fix: event classifier now strips markdown fences and repairs JSON ...

_parse_classification_response receives raw model output (with thinking
tags, markdown fences, etc.) but was calling json.loads directly.
Now uses _strip_markdown_fences + _repair_json from the client module
before parsing, matching what _call_ollama does for extractions.

2026-04-17 16:35:13 +00:00

..

__init__.py

phase 0+1: project scaffold, k8s manifests, CI pipeline, steering, hooks, tests

2026-04-11 03:25:08 -07:00

client.py

feat: agent variants — migration, API, service integration, frontend, tests

2026-04-17 05:15:42 +00:00

event_classifier.py

fix: event classifier now strips markdown fences and repairs JSON

2026-04-17 16:35:13 +00:00

exposure_inference.py

feat: competitive intelligence & historical pattern matching layer

2026-04-14 19:42:48 +00:00

main.py

feat: wire all 3 agents to DB config resolver

2026-04-17 02:59:40 +00:00

metrics.py

phase 14-15: docker build validation and helm deployment

2026-04-11 11:59:45 -07:00

prompts.py

fix: reduce LLM timeouts — truncate docs to 8k/6k chars, cut num_predict 16k→4k, tighten prompts, trim anti-hallucination rules

2026-04-16 18:56:11 +00:00

replay.py

phase 14-15: docker build validation and helm deployment

2026-04-11 11:59:45 -07:00

schemas.py

fix: fill default values for missing fields in truncated LLM output

2026-04-15 03:08:10 +00:00

worker.py

phase 15: fix ruff lint errors across services

2026-04-11 12:10:01 -07:00