Remove the rule-advisor lineage and reduce the diagnosis entry - #179
Merged
Merged
Conversation
rule-advisor selected skills and returned section names, but recipe-task read each selected skill completely, so the selection round trip saved no context. Skill descriptions already drive that selection, and the metacognitive warnings duplicated ai-development-guide, coding-principles, and testing-principles. Remove the agent, task-analyzer, recipe-task, and the skills index whose only consumer was that lineage, along with the pre-commit check that kept the index in sync. recipe-diagnose consumed rule-advisor in Step 0.3 and stopped for AskUserQuestion whenever the change type or change details were unclear. Resolve those from the report and repository evidence, carry what stays unresolved into the investigator prompt as investigation targets, and renumber the scope envelope to 0.3. Drop the two enforcement restatements in recipe-implement and recipe-fullstack-implement; each repeated the numbered flow directly above it without adding an observable condition. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Step 0 classified the problem type and collected change details to decide whether the investigator prompt carried three extra fields. Nothing downstream consumed the type, and the scope envelope already records the phenomenon, its occurrence conditions, and history as an evidence axis. Remove the classification, the supplementation block, and the conditional prompt fields, leaving the envelope as Step 0. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
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.
Changes
Removed
agents/rule-advisor.mdskills/task-analyzer/(includingreferences/skills-index.yaml)skills/recipe-task/scripts/check-skills-index.mjsand the lefthook job and npm script that only ran itrule-advisor returned skill and section names, but
recipe-taskthen read each selected skill completely, so the round trip saved no context.investigator,verifier, andsolveralready declareai-development-guideandcoding-principlesin their own frontmatter, so the rules reached them without the agent.recipe-diagnose
AskUserQuestiongates that stopped diagnosis before investigation began.Other prompts
**CRITICAL**restatement inrecipe-implementandrecipe-fullstack-implement. Each repeated the numbered flow directly above it, stop points included.Docs and version
/recipe-taskandrule-advisorrows from the six READMEs.Verification
pnpm sync:checkin sync.claude plugin validatepasses for the marketplace manifest and all four plugins. No references torule-advisor,task-analyzer,recipe-task, orskills-indexremain.Net: 26 insertions, 1992 deletions.
🤖 Generated with Claude Code