feat(sdlc): graduate confidence ramp + advisor fallback patterns#412
Merged
Conversation
Confidence ramp pattern (Opus research → Fable batch review → 95% list → TDD) proven on v1.83.0 model-config batch. Advisor auto-fallback (spawn Fable subagent when advisor() unavailable) proven across 3 sessions. Both now codified in skills/sdlc/SKILL.md. Trimmed CI log audit + claude-md-improver refs to stay under 20K budget. 6 TDD tests cover pattern presence, batch review step, 95% gate, fallback documentation, Fable subagent instruction, and char budget.
BaseInfinity
force-pushed
the
feat/sdlc-skill-graduations
branch
from
e0a1cab to
b3668bb
Compare
Round 1 NOT CERTIFIED (3 P1s): - Ramp text now includes /goal + Codex check (was incomplete) - Advisor fallback moved near Plan Mode (was only in Cross-Model Review) - Test 3 scoped to confidence ramp line (was matching pre-existing /goal)
BaseInfinity
force-pushed
the
feat/sdlc-skill-graduations
branch
from
b3668bb to
9d096f6
Compare
3 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Graduates two proven patterns from the Research Parking Lot to the SDLC skill:
Confidence ramp (Confidence Check section): Opus researches → batch-consult Fable advisor → build 95%+ list → TDD each. Proven on v1.83.0 model-config batch (4 issues, Codex certified).
Advisor auto-fallback (Cross-Model Review section): When
advisor()is unavailable, spawn a Fable subagent as the planning reviewer. Proven across 3 sessions.Also trimmed CI log audit reference (redundant with §3) and claude-md-improver line to stay under 20K budget.
Fable review
Fable triaged the ROADMAP and identified both items as Priority 1 and 2 — triggers fired, ready to graduate.
Test plan
tests/test-skill-graduations.sh— 6/6 passing (TDD RED→GREEN)tests/test-doc-consistency.sh— 42/42 (no regression)