review-agent-corrections

Review locally captured corrections to agent edits and propose reusable repository guidance. Use when asked to review agent corrections, learn from feedback, inspect pending correction events, or improve future agent edits from prior feedback.

Quality

91%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

100%Weight 40%Scale 1-5

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

The body is an exemplary instruction-only skill: lean, fully actionable with exact commands and target files, a clearly sequenced workflow with explicit approval and per-event disposition checkpoints, and well-organized sections requiring no external references.

Dimension	Reasoning	Score
Conciseness	The body is lean with no padding or explanation of concepts Claude already knows; every section and line earns its place, matching the score-5 anchor for lean, efficient content.	5 / 5
Actionability	Provides fully executable, copy-paste-ready commands with exact arguments (e.g. 'node .github/hooks/track-agent-feedback.js pending' and acknowledge/dismiss/defer with event IDs) plus exact target files and concrete classification buckets, matching the score-5 anchor.	5 / 5
Workflow Clarity	Clear sequenced multi-step process with explicit validation/approval checkpoints ('If there are no pending events... stop', 'Stop and wait for explicit approval', 'After approval:') and per-event disposition, matching the score-5 anchor for explicit validation steps and feedback loops.	5 / 5
Progressive Disclosure	Self-contained skill with no bundle files; content is compact and cleanly organized into well-signaled sections (Gather evidence, Classify and correlate, Deduplicate, Get approval, Apply the decision), meeting the simple-skill exception for a 5 with well-organized sections.	5 / 5
	Total	20 / 20 Passed

Description

82%Weight 40%Scale 1-5

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 third-person, complete with explicit 'what' and 'when' triggers, and occupies a distinct niche. The main weakness is specificity: it names only two high-level actions rather than enumerating a comprehensive concrete action set.

Suggestions

Expand the action list to be more comprehensive, e.g. 'Review locally captured corrections, classify them, correlate related events, deduplicate signals, and propose reusable repository guidance or deterministic hooks.'

Add file-type or artifact synonyms (e.g. 'diffs', 'prompts', 'agent feedback events') to broaden natural trigger coverage toward the score-5 anchor.

Dimension	Reasoning	Score
Specificity	Names the domain ('corrections to agent edits') and two concrete actions ('Review... corrections', 'propose reusable repository guidance'), but does not enumerate a comprehensive set of specific actions, matching the anchor for domain plus 1-2 actions.	3 / 5
Completeness	Explicitly answers both what ('Review... corrections and propose reusable repository guidance') and when via a concrete 'Use when...' clause listing multiple trigger scenarios, matching the score-5 anchor.	5 / 5
Trigger Term Quality	Provides several natural phrases users would say ('review agent corrections', 'learn from feedback', 'inspect pending correction events', 'improve future agent edits'), giving good keyword coverage with a few common synonyms missing, fitting the score-4 anchor rather than the comprehensive score-5 anchor.	4 / 5
Distinctiveness Conflict Risk	Targets a clear niche (locally captured corrections to agent edits, pending correction events) with distinct triggers and minimal overlap with other skills, matching the score-5 anchor.	5 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: microsoft/vscode-docs
Path: .github/skills/review-agent-corrections/SKILL.md
Commit: 193d3e4

Reviewed: 3 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.