REPO_STATUS: part-2 header entry; lifecycle notes pre-plan retrieval; feature
list corrected to 4 LLM calls (guard counted) + retrieve-then-plan; Langfuse
section records the E2 planner masking. DEV_PLAN: E1 decision + T3 recheck,
new P2-G1..G3 (eval records, call labels, fragile chart case). Plan §8 on eval
records; docs index marks A-E done.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
[NOTICKET] docs: retrieve-then-plan relevance cut stays at 0.5 (decided)
Rifqi Hafizuddin+1: Claude Opus 5.5
55a0ec2
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
eval: mixed T3 re-checked 3x with full records; label calls by output
Rifqi Hafizuddin+1: Claude Opus 5.5
1cd7221
T3 ("according to the standard...") x3 via --turns 3: 2/3 plans add
retrieve_knowledge, 1/3 answer from the approved domain card (stated in its
assumptions); identical correct site figures; no retries. Retrieve-then-plan
kept nothing in 3/3 (EN question vs ID standard: 0.584-0.643 > 0.5 cut).
chat_sim labelled LLM calls by order, so the input-guard call shifted every
label by one; labels now come from each call's output fields. The three new
transcripts were relabelled from their saved outputs (outputs untouched).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
[NOTICKET] docs: pr/23 part 2 D + E build record, traceability contract note
Rifqi Hafizuddin+1: Claude Opus 5.5
153b12f
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
feat(router): intents defined once, 21 few-shots, sharper group_hint groups
Rifqi Hafizuddin+1: Claude Opus 5.5
1dfbf6d
C1: drop the restating Routing rules section. C2: 11 definition-restating
few-shots removed (kept as eval rows); explore/grouping/evaluate sharpened.
eval.intent 57/57 both runs; group_hint 12 exact / 3 superset / 0 wrong;
router prompt 3,484 -> 3,164 tokens. Scorer gains a superset bucket.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
[NOTICKET] docs: pr/23 part 2 plan re-assessed, A + B build record
Rifqi Hafizuddin+1: Claude Opus 5.5
415c4ef
- docs/plans/PR23_PART2_PLANNER_EVAL.md: the older C/L/D items re-checked
against the code, sections A-F, build record, A7 audit (exemplars reach
the prompt without their explanations; compact JSON = new D5, -35%).
- DEV_PLAN §0.11 task rows; REPO_STATUS query engine / planner / assembler
notes; dated corrections in PR2_EVALUATION, SPINE_V2 (4 LLM calls) and the
prompt-architecture review.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
fix(chatbot): reply in the turn's language, not the documents'
Rifqi Hafizuddin+1: Claude Opus 5.5
b6150d5
An English question about an Indonesian document was answered in Indonesian.
detect_reply_language already returned English; the model followed the
Indonesian excerpts, and the rule only warned about English context pulling
the reply. The rule is now language-neutral: this turn's language wins
(explicit "answer in X" still overrides).
Doc smoke (turns 2,4): ID -> ID, EN -> EN (transcript_doc_2026-09-23_141212.json,
~14k). 65 related tests pass. Docs: DEV_PLAN §6 (c), REPO_STATUS §9, plan §4.4.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- UNSTRUCTURED_V2_PLAN §4.5: the mixed run (T2/T3 one plan with document +
data, T4 numbers-only without the document, T1 check bug -> fixed) and the
router fix with its phrasing-rule caveat.
- DEV_PLAN §0.10 U8 done, U9 drafted; §6 follow-ups (a) and (d) resolved.
- REPO_STATUS §9: router + check behaviour, known limits.
- transcript_mixed_2026-09-23_132033.json (run by Rifqi after the catalog rebuild).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
[NOTICKET] docs: unstructured v2 plan, contract, status sync
Rifqi Hafizuddin+1: Claude Opus 5.5
6702119
- docs/plans/UNSTRUCTURED_V2_PLAN.md: the 22/9 flow, decisions U-D1..U-D7,
scope audit, U0 findings, build record U1-U8.
- API contract: Documents (v2) section — 5 routes, delete stays on Go,
retrieval-scope behaviour change.
- REPO_STATUS §4/§6/§7/§8/§9/§12/§13 + header delta; DEV_PLAN §0.10 + §6.
- docs/reviews/PROMPT_ARCHITECTURE_2026-09-22.md (planner + router assessment,
paused behind the demo track) and a reviews/ section in the docs index.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
