learn-from-pr

Analyzes a completed PR to extract lessons learned from agent behavior. Use after any PR with agent involvement - whether the agent failed, succeeded slowly, or succeeded quickly. Identifies patterns to reinforce or fix, and generates actionable recommendations for instruction files, skills, and documentation.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./.github/skills/learn-from-pr/SKILL.md

Quality

Content

65%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is well-structured and highly actionable, with concrete commands, a precise recommendation schema, and strong worked examples. Its weaknesses are token efficiency (some exhaustive inline lists could be trimmed or split out) and workflow clarity (validation/recovery is described separately rather than woven in as explicit in-flow checkpoints).

Suggestions

Trim or consolidate the longest enumerative tables (the 10-row 'Agent Failed' pattern list and the pattern-to-improvement mappings) to the highest-value rows, moving the full catalog to a reference file.

Integrate error handling as inline checkpoints within the 6-step workflow (e.g., 'Step 1 — if gh pr view fails, stop and ask the user to verify the PR number') rather than only in a separate table.

Split the three full worked examples into an examples/ reference and keep only a brief pointer plus one condensed example in SKILL.md to improve progressive disclosure.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient — heavy use of tables, checklists, and concise sections with no explaining of concepts Claude already knows — but several exhaustive enumerative tables (e.g., the 10-row and 4-row scenario/pattern lists) and repeated 'what happened / what made it hard' framing add padding an expert Claude could partly infer.	2 / 3
Actionability	Provides concrete commands ('gh pr view XXXXX --json ...', 'gh pr diff XXXXX', find commands), a precise recommendation schema (Category/Priority/Location/Specific Change/Why It Helps), and fully-worked example PRs with exact file paths and exact 'Change' text, giving copy-paste-ready guidance for an instruction skill.	3 / 3
Workflow Clarity	There is a clear numbered 6-step sequence plus a Completion Criteria checklist, but validation checkpoints are implicit rather than embedded in the flow — error/recovery guidance lives in a separate 'Error Handling' table instead of integrated 'if X, then stop and ask' checkpoints within the steps.	2 / 3
Progressive Disclosure	The skill is a single ~270-line file with no bundle references, and while it is well sectioned, the exhaustive inline pattern tables and three full worked examples are content that a larger skill would typically split into one-level-deep reference files.	2 / 3
	Total	9 / 12 Passed

Description

85%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and complete, clearly stating what the skill does and when to use it with an explicit 'Use after' trigger. Its main weakness is trigger-term coverage: scenario phrasings are present but the most common user phrasings ('learn from this PR', 'what went wrong with PR #X') are absent.

Suggestions

Add common natural trigger phrases such as 'Use when the user asks to learn from PR #X, review what an agent did wrong, or extract lessons learned from a completed PR'.

Lead with the highest-frequency user phrasing (e.g., 'Extract lessons learned from a completed PR') before the scenario list so the most common trigger term appears up front.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Analyzes a completed PR to extract lessons learned', 'Identifies patterns to reinforce or fix', and 'generates actionable recommendations for instruction files, skills, and documentation' — matching the multiple-specific-actions anchor.	3 / 3
Completeness	Explicitly answers both what ('Analyzes a completed PR...extracts lessons learned...generates actionable recommendations') and when (explicit 'Use after any PR with agent involvement' clause), satisfying the both-what-and-when anchor.	3 / 3
Trigger Term Quality	Has scenario-based triggers ('Use after any PR with agent involvement - whether the agent failed, succeeded slowly, or succeeded quickly') but misses common natural phrasings a user would say like 'lessons learned from this PR' or 'what went wrong with PR #X', so it does not reach full natural-term coverage.	2 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (post-hoc lessons-learned analysis of agent-involved PRs) with distinct triggers, and is differentiated from pr-finalize in the body, making conflict with other skills unlikely.	3 / 3
	Total	11 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: dotnet/maui
Path: .github/skills/learn-from-pr/SKILL.md
Commit: 6a24ec5

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

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.