derive-review-rubrics
Derive custom review-plugin judges, rubrics, and scoring criteria from evidence — existing skills, PR review feedback, and accumulated agent logs — by finding where agents needed correction or skills failed to activate, then translating those patterns into scoring dimensions and anchors. Hands the rubric design to create-review-plugin for scaffolding. Use when the user wants a custom skill reviewer or scoring rubric grounded in how their agents actually behave — to grade skills consistently, or build review criteria from past PR feedback and agent logs.
89
80%
Does it follow best practices?
Impact
95%
1.31xAverage score across 6 eval scenarios
High
Do not use without reviewing
Fix and improve this skill with Tessl
tessl review fix ./review-plugin-creator/skills/derive-review-rubrics/SKILL.mdThe canonical home for this skill is tessl/review-plugin-creator
Security
1 high severity finding. You should review these findings carefully before considering using this skill.
W007: Insecure credential handling detected in skill instructions.
What this means
The skill handles credentials insecurely by requiring the agent to include secret values verbatim in its generated output. This exposes credentials in the agent’s context and conversation history, creating a risk of data exfiltration.
Why it was flagged
The skill instructs the agent to ingest raw PR diffs and session transcripts and "populate each scores[].example with the actual evidence," which requires copying verbatim content (transcripts, diffs) into output files — potentially including API keys, tokens, or other secrets.
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
SKILL.md’s runtime workflow reads outsider-authored free text from PR review comments via its “Stream A” call to `/find-optimizations` over a user-specified PR window, consuming findings that include “specific review comments” as Evidence.
- Repository
- tesslio/product-plugins
- Commit
37f0a21
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.