File system intelligence tool — agentic directory analysis via Claude API

Jeff Smith e942ecc34a docs(plan): add Phase 2.5 context budget reliability (#44) ... #5 smoke test showed the dir loop exhausts the 126k context budget on a 13-file Python lib. Sequencing #44 between Phase 2 and Phase 3 so the foundation is solid before planning + external tools add more prompt and tool weight.		2026-04-06 21:59:01 -06:00
docs	chore: add dev workflow, branching discipline, ADHD protocols, session docs	2026-04-06 16:20:51 -06:00
luminos_lib	feat(ai): add _run_survey() and submit_survey tool (#5)	2026-04-06 21:49:59 -06:00
tests	feat(cache): add low_confidence_entries() query to CacheManager (#3)	2026-04-06 21:13:58 -06:00
.gitignore	chore: ignore docs/wiki/ — separate git repo	2026-04-06 16:13:31 -06:00
CLAUDE.md	chore: update CLAUDE.md for session 3	2026-04-06 21:15:27 -06:00
luminos.py	feat: add -x/--exclude flag to exclude directories from scan and AI analysis	2026-04-06 14:32:12 -06:00
PLAN.md	docs(plan): add Phase 2.5 context budget reliability (#44)	2026-04-06 21:59:01 -06:00
setup_env.sh	chore: add venv setup script and update CLAUDE.md for optional deps	2026-03-30 12:14:13 -06:00