improve consensus skill description + inline workflow#23
Merged
cofin merged 1 commit intocofin:mainfrom Apr 3, 2026
Merged
Conversation
- expand description with concrete outputs + trigger terms - inline stance rotation steps + synthesis framework - add validation checkpoint + concrete example
Owner
|
@fernandezbaptiste thanks for the review and for checking out the skills! This info helps, and I'll check out |
cofin
approved these changes
Apr 3, 2026
cofin
added a commit
that referenced
this pull request
Apr 3, 2026
…skills Systematically improve all skills based on the consensus skill improvements from PR #23 (69% → 94% agent performance). Changes across three tiers: Full treatment (25 skills): description enrichment with output shape + explicit exclusions, inlined workflow steps from references, <guardrails> constraints, <validation> checkpoints, concrete <example> sections. Covers all persona skills (devils-advocate, architecture-critic, performance-analyst, security-auditor), process skills (challenge, deepthink, tracer, docgen), and technology skills that fully delegated to references. Medium treatment (13 skills): description enrichment, semantic tags, minor structural improvements. Light treatment (17 skills): semantic tags on existing well-structured content. Flow skill: added validation checkpoints for TDD workflow and phase completion protocol, semantic tags on workflow sections, description exclusions. Introduces a 4-tag semantic XML standard (<workflow>, <guardrails>, <validation>, <example>) to help agents distinguish instructions from context. Moved markdownlint config to project root with allowed elements for the new tags. Fixed all 106 markdownlint issues across 292 files.
Contributor
Author
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
2 participants
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.
hey @cofin, thanks for building flow. really like the "measure twice, code once" philosophy + the breadth of 55 technology-specific skills. kudos for the repo! just starred it.
ran your consensus skill through some evals and noticed a few things that were pretty quick to improve (moving from
~69%to~94%agent performance):these were easy changes to bring the skill in line with what performs well against Anthropic's best practices. honest disclosure, I work at tessl.io where we build tooling around this. not a pitch, just fixes that were straightforward to make.
you've got
55skills, if you want to do it yourself, spin up Claude Code and runtessl skill review. alternatively, let me know if you'd like an automatic review in your repo via GitHub Actions. it doesn't require signup, and this means you and your contributors get an instant quality signal before you have to review yourself.