common-session-retrospective
Analyze conversation corrections to detect skill gaps and auto-improve the skills library. Use after any session with user corrections, rework, or retrospective requests. After finding correction loops, also load +common/common-learning-log to persist mistake entries to AGENTS_LEARNING.md. (triggers: **/*.spec.ts, **/*.test.ts, SKILL.md, AGENTS.md, retrospective, self-learning, improve skills, session review, correction, rework, +common/common-learning-log)
90
88%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
92%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong description that clearly communicates what the skill does and when to use it, with an explicit 'Use after...' clause and a comprehensive triggers list. The main weakness is that some triggers (test file globs, SKILL.md, AGENTS.md) are quite broad and could cause conflicts with other skills focused on testing or skill/agent file management. The description uses proper third-person voice throughout.
Suggestions
Narrow the file-pattern triggers (e.g., **/*.spec.ts, **/*.test.ts) or clarify that these only apply in the context of correction/rework analysis, to reduce overlap with testing-focused skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions: 'analyze conversation corrections', 'detect skill gaps', 'auto-improve the skills library', 'persist mistake entries to AGENTS_LEARNING.md', and 'load +common/common-learning-log'. These are specific, actionable capabilities. | 3 / 3 |
Completeness | Clearly answers 'what' (analyze corrections, detect skill gaps, auto-improve skills library, persist learning entries) and 'when' ('Use after any session with user corrections, rework, or retrospective requests'). The explicit 'Use after...' clause and triggers list make this fully complete. | 3 / 3 |
Trigger Term Quality | Includes an explicit triggers list with natural terms users would say: 'retrospective', 'self-learning', 'improve skills', 'session review', 'correction', 'rework', plus relevant file patterns like '**/*.spec.ts', 'SKILL.md', 'AGENTS.md'. Good coverage of both natural language and file-based triggers. | 3 / 3 |
Distinctiveness Conflict Risk | While the core concept of analyzing corrections and improving skills is fairly distinct, the broad file triggers like '**/*.spec.ts' and '**/*.test.ts' could easily conflict with testing-related skills. The trigger 'SKILL.md' and 'AGENTS.md' could also overlap with skills that manage or edit those files for other purposes. | 2 / 3 |
Total | 11 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured meta-skill with excellent progressive disclosure and workflow clarity. The protocol is concise and well-sequenced with clear references to supporting materials. The main weakness is that actionability could be improved with concrete examples of correction signal detection, classification output, or a mini example of a trigger miss block inline rather than solely deferring to the reference file.
Suggestions
Add a brief inline example of a correction signal detection and its classification (e.g., a concrete before/after showing a skill gap identified from a conversation excerpt) to improve actionability.
Include a minimal inline example of the trigger miss output block rather than only referencing the schema in methodology.md, so Claude can act without needing to load the reference file for simple cases.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every section is lean and purposeful. No unnecessary explanations of concepts Claude already knows. The protocol steps are terse but clear, and anti-patterns are stated as brief rules rather than verbose explanations. | 3 / 3 |
Actionability | The protocol provides a clear numbered sequence of actions, but guidance remains somewhat abstract — there are no concrete code examples, specific commands, or copy-paste-ready templates for extraction or classification. The actual signal tables and report template are deferred to references/methodology.md. | 2 / 3 |
Workflow Clarity | The 7-step protocol is clearly sequenced with logical progression from extraction through classification, proposal, implementation, logging, and reporting. The Trigger Miss Check (step 3) acts as an explicit validation checkpoint, and the logging step (step 6) provides a feedback/persistence mechanism. | 3 / 3 |
Progressive Disclosure | The file structure is shown upfront, the SKILL.md stays concise as an overview, and detailed content (signal tables, taxonomy, report template, trigger miss schema) is clearly referenced via one-level-deep links to references/methodology.md with specific anchors. | 3 / 3 |
Total | 11 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- HoangNguyen0403/agent-skills-standard
- Commit
19a1140
- Reviewed
Table of Contents
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.