feat: improve skill scores for autonomous-dev#901
Open
yogesh-tessl wants to merge 1 commit into
Open
Conversation
Hey @akaszubski 👋 I ran your skills through `tessl skill review` at work and found some targeted improvements. Here's the full before/after: | Skill | Before | After | Change | |-------|--------|-------|--------| | api-design | 74% | 83% | +9% | | python-standards | 82% | 89% | +7% | | code-review | 90% | 93% | +3% | | debugging-workflow | 90% | 90% | — | | testing-guide | 86% | 86% | — |  **Note:** All five skills had their `allowed-tools` frontmatter field changed from YAML array syntax (`[Read]`) to the required string format (`"Read"`), which was blocking the LLM judge from scoring them. The "before" scores above reflect scores *after* that format fix, so the improvement numbers reflect genuine content improvements only. ## Type of Change - [x] Refactor (code improvement, no behavior change) ## Changes <details> <summary>What changed and why</summary> ### api-design (+9%) - Removed redundant "When This Skill Activates" section (duplicates frontmatter description triggers) - Removed HTTP status codes listing (Claude already knows standard codes) - Removed "Key Takeaways" section (duplicated Core Concepts) - Removed "Progressive Disclosure" meta-section explaining the file's own structure - Consolidated "Quick Reference" table and "Available Documentation" list into a single "Reference Documentation" table - Added sequenced "API Design Workflow" section (8-step process) to address missing workflow clarity - Fixed error format inconsistency in Hard Rules (`{error, message, code}` → RFC 7807 to match the Error Handling section) - Added file paths to cross-references for navigability - Removed "Related Libraries" section (generic knowledge) ### python-standards (+7%) - Improved description specificity with concrete action verbs ("Enforces PEP 8 compliance, applies Black formatting...") - Removed redundant "When This Activates" section (duplicates frontmatter) - Removed "Key Takeaways" section (duplicated earlier content) - Consolidated duplicate error message format documentation - Added sequenced "Quality Check Workflow" with numbered steps (format → sort → lint → type check → test) - Added file paths to cross-references ### code-review (+3%) - Deduplicated anti-patterns section (removed entries that restated HARD GATE rules) - Added actual file paths to cross-references (`../python-standards/SKILL.md` instead of bare names) ### debugging-workflow (0%) - Replaced redundant "Common Pitfalls" table (duplicated guidance from Phase 1 anti-patterns and Phase 4 rules) with cross-references to related skills ### testing-guide (0%) - Fixed `allowed-tools` frontmatter format only (content was already strong at 86%) ### All skills - Fixed `allowed-tools` frontmatter from YAML array syntax (`[Read]` / `[Read, Grep, Glob, Bash]`) to required string format (`"Read"` / `"Read, Grep, Glob, Bash"`) </details> ## Test Plan - [x] All skills pass `tessl skill review` validation (0 errors) - [x] Total diff under 300 lines (297 lines) - [x] No behavioral changes — skills provide the same guidance, just more concisely ## Quality Checklist - [x] Code follows project standards - [x] Documentation updated (skill content improved) - [x] No security issues detected - [x] Commit messages follow conventional format - [x] Aligns with PROJECT.md goals I kept this PR focused on the 3 skills with the biggest improvements to keep the diff reviewable. Happy to follow up with the rest in a separate PR if you'd like. Honest disclosure — I work at @tesslio where we build tooling around skills like these. Not a pitch — just saw room for improvement and wanted to contribute. Want to self-improve your skills? Just point your agent (Claude Code, Codex, etc.) at [this Tessl guide](https://docs.tessl.io/evaluate/optimize-a-skill-using-best-practices) and ask it to optimize your skill. Ping me — [@yogesh-tessl](https://github.com/yogesh-tessl) — if you hit any snags. Thanks in advance 🙏
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.
Hey @akaszubski 👋
An 8-agent pipeline that enforces alignment checks, adversarial code review, and security audits before a single line ships, v3.51 with 1,000+ commits shows this isn't a weekend experiment. The PROJECT.md-as-contract pattern is clever. Wanted to suggest a few improvements to the SKILL.md.
I ran your skills through
tessl skill reviewat work and found some targeted improvements. Here's the full before/after:What changed and why
api-design (+9%)
{error, message, code}→ RFC 7807 to match the Error Handling section)python-standards (+7%)
code-review (+3%)
../python-standards/SKILL.mdinstead of bare names)debugging-workflow (0%)
testing-guide (0%)
allowed-toolsfrontmatter format only (content was already strong at 86%)All skills
allowed-toolsfrontmatter from YAML array syntax ([Read]/[Read, Grep, Glob, Bash]) to required string format ("Read"/"Read, Grep, Glob, Bash")Honest disclosure. I work at https://github.com/tesslio where we build tooling around skills like these. Not a pitch - just saw room for improvement and wanted to contribute.
Want to self-improve your skills? Just point your agent (Claude Code, Codex, etc.) at this Tessl guide (https://docs.tessl.io/evaluate/optimize-a-skill-using-best-practices) and ask it to optimize your skill. Ping me - @yogesh-tessl (https://github.com/yogesh-tessl) - if you hit any snags.
Thanks in advance 🙏