feat: fix skill frontmatter validation across 22 skills#16
feat: fix skill frontmatter validation across 22 skills#16popey wants to merge 1 commit intoteam-attention:mainfrom
Conversation
Hullo @yansfil 👋 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 | |-------|--------|-------|--------| | compound | 0% | 95% | +95% | | council | 0% | 89% | +89% | | specify | 0% | 89% | +89% | | stepback | 0% | 89% | +89% | | execute | 0% | 88% | +88% | | ralph | 0% | 84% | +84% | | rulph | 0% | 80% | +80% | | skill-session-analyzer | 0% | 80% | +80% | | tech-decision | 0% | 79% | +79% | | deep-research | 0% | 70% | +70% | | ultrawork | 0% | 68% | +68% | | deep-interview | 0% | 57% | +57% | | discuss | 0% | 57% | +57% | | mirror | 0% | 57% | +57% | | tribunal | 79% | 95% | +16% | | quick-plan | 50% | 57% | +7% | | google-search | 94% | 100% | +6% | | check | 88% | 93% | +5% | | browser-work | 84% | 84% | — | | dev-scan | 84% | 84% | — | | issue | 100% | 100% | — | | scope | 96% | 96% | — | | bugfix | 64% | 55% | -9% | | reference-seek | 80% | 75% | -5% | The 0% "before" scores were caused by deterministic validation failures — the review couldn't even get to the LLM judge. Fixing those unlocked proper scoring. Changes made: - allowed-tools format fix (14 skills) — The allowed-tools frontmatter field was declared as a YAML array (list) instead of the expected comma-separated string. This caused deterministic validation failure, blocking the LLM judge entirely and producing 0% scores. Affected: compound, council, deep-interview, discuss, execute, mirror, ralph, rulph, skill-session-analyzer, specify, stepback, tribunal, ultrawork, quick-plan. - allowed_tools key fix (quick-plan) — Used underscore (allowed_tools) instead of hyphen (allowed-tools), which is the standard frontmatter key. - Description format normalization (22 skills) — Converted | (literal block scalar) and > (folded block scalar) YAML syntax to standard quoted strings for the description field. - XML tag removal (deep-research) — Description contained <topic> which failed the "description must not contain XML tags" validation. Replaced with [topic]. 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 - @popey (https://github.com/popey) - if you hit any snags. Thanks in advance 🙏
Summary of ChangesHello, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request enhances the robustness and evaluability of 22 skill definitions by resolving critical frontmatter validation errors. These changes, identified through Highlights
🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console. Changelog
Activity
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request effectively addresses the frontmatter validation issues across 22 skills as described. The changes correctly normalize the description fields from YAML block scalars to standard quoted strings and convert allowed-tools (or allowed_tools) from YAML arrays to comma-separated strings. These adjustments align with the stated goal of fixing deterministic validation failures and improving skill scoring. The changes are well-executed and do not introduce any new issues.
1 participant
Hullo @yansfil 👋
I ran your skills through
tessl skill reviewat work and found some targeted improvements. Here's the ten most improved:Here's the full before/after in text form:
The 0% "before" scores were caused by deterministic validation failures — the review couldn't even get to the LLM judge. Fixing those unlocked proper scoring.
Changes made:
allowed-tools format fix (14 skills) — The allowed-tools frontmatter field was declared as a YAML array (list) instead of the expected comma-separated string. This caused deterministic validation failure, blocking the LLM judge entirely and producing 0% scores. Affected: compound, council, deep-interview, discuss, execute, mirror, ralph, rulph, skill-session-analyzer, specify, stepback, tribunal, ultrawork, quick-plan.
allowed_tools key fix (quick-plan) — Used underscore (allowed_tools) instead of hyphen (allowed-tools), which is the standard frontmatter key.
Description format normalization (22 skills) — Converted | (literal block scalar) and > (folded block scalar) YAML syntax to standard quoted strings for the description field.
XML tag removal (deep-research) — Description contained which failed the "description must not contain XML tags" validation. Replaced with [topic].
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 and ask it to optimize your skill. Ping me - @popey - if you hit any snags.
Thanks in advance 🙏