Colby McHenry 2cfd321b23 docs: CG-20 — the dedup gate, three bars pass and the fourth cannot (CG-2) ...

Read is 0 in all 24 runs of both arms across client-go and excalidraw, no
isError, codegraph is the last tool in every run, and both "Read a file we
returned" / "did not return" buckets are empty — with back-references
demonstrably reaching the agent in 8 of the 9 multi-call runs.

Residual occupancy is flat, and the measurement shows it could not have been
anything else: CG-18 was accepted on the rule that reclaimed bytes get spent on
unseen files rather than banked, so the byte count cannot fall. What moves is
the duplicate fraction of that residual — 87% less across the agent runs, 86%
and 94% on two matched deterministic replays.

Also recorded: dedup.savedChars is a pre-clip figure (11,450 reported against
1,042 chars actually re-served), so tuning off it inflates the win ~7x.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

2026-08-05 14:11:33 -05:00

..

answer-directly-vs-explore-agent.md

docs(readme): answer directly with codegraph, not via an Explore agent (#367)

2026-05-24 04:52:44 -05:00

call-sequence-analysis.md

Release 0.9.4: framework-aware routing + dynamic-dispatch coverage + retrieval improvements (#365)

2026-05-24 04:41:04 -05:00

codegraph-ab-matrix.md

Add landing page + Starlight docs site (#375)

2026-05-24 13:21:25 -05:00

explore-allocation-ab-1500.md

docs: CG-22 — the epic's gate, re-run at CG-15's exact setup (#1500)

2026-08-04 12:58:26 -05:00

explore-dedup-ab-cg20.md

docs: CG-20 — the dedup gate, three bars pass and the fourth cannot (CG-2)

2026-08-05 14:11:33 -05:00