Append a structured learning entry to AGENTS_LEARNING.md whenever an AI agent makes a mistake. Auto-activates as a composite skill when: a pre-write skill violation is detected and auto-fixed, or when the session retrospective finds a correction loop. Also triggers directly when the user corrects the AI mid-session. Use when: mistake, wrong, redo, that's not right, correction, my bad, fix that error, I made a mistake, agent error, learning log, log mistake, AGENTS_LEARNING.md
71
88%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
100%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 an excellent skill description that clearly defines a specific niche (logging AI agent mistakes to AGENTS_LEARNING.md), provides comprehensive trigger conditions including both automatic and manual activation scenarios, and includes a rich set of natural trigger terms. The description is well-structured, uses third person voice appropriately, and would be easily distinguishable from other skills in a large skill library.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description lists multiple specific concrete actions: 'Append a structured learning entry to AGENTS_LEARNING.md', 'pre-write skill violation is detected and auto-fixed', 'session retrospective finds a correction loop', and 'user corrects the AI mid-session'. These are detailed, concrete behaviors. | 3 / 3 |
Completeness | Clearly answers both 'what' (append structured learning entries to AGENTS_LEARNING.md when mistakes occur) and 'when' (explicit 'Use when:' clause with trigger terms, plus detailed auto-activation conditions). Both dimensions are thoroughly covered. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms users would say: 'mistake', 'wrong', 'redo', 'that's not right', 'correction', 'my bad', 'fix that error', 'I made a mistake', 'agent error', 'learning log', 'log mistake', and the specific file 'AGENTS_LEARNING.md'. These are highly natural phrases. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche: logging AI agent mistakes to a specific file (AGENTS_LEARNING.md). The combination of error correction, learning logs, and the specific file target makes it very unlikely to conflict with other skills like general error handling or code fixing. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%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, concise skill with a clear workflow for logging agent mistakes. Its main weaknesses are the broken reference to `references/log-format.md` (not provided in the bundle) and the lack of any inline example showing what a completed log entry looks like. There are also minor typos in the Guidelines section ('what to' instead of 'what to do', 'all corrections learning signals' missing 'are').
Suggestions
Either provide the referenced `references/log-format.md` bundle file or include an inline example of a completed log entry so the skill is self-contained and actionable.
Fix typos in Guidelines: '...state what to [do], not what to avoid' and '...all corrections [are] learning signals'.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is lean and efficient. Every section serves a purpose—protocol steps, guidelines, and anti-patterns are all tightly written without explaining concepts Claude already knows. No unnecessary preamble or padding. | 3 / 3 |
Actionability | The protocol steps are clear and specific (read file, count headers, append entry), but the actual log entry format is deferred to a reference file that isn't provided in the bundle. The skill lacks a concrete inline example of what a completed entry looks like, making it incomplete for copy-paste execution. | 2 / 3 |
Workflow Clarity | The 4-step protocol is clearly sequenced with explicit trigger detection, state reading, writing, and continuation. The guidelines add important constraints (one entry per event, create file if missing, never skip). The non-destructive append-only nature reduces the need for validation checkpoints, and the workflow is unambiguous. | 3 / 3 |
Progressive Disclosure | The skill references `references/log-format.md` for the entry template and bootstrap header, which is good progressive disclosure structure. However, the bundle files show no such file exists, meaning the reference is broken. The skill would benefit from either including the template inline or actually providing the referenced file. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
metadata_version | 'metadata.version' is missing | Warning |
metadata_field | 'metadata' should map string keys to string values | Warning |
Total | 9 / 11 Passed | |
3df717f
Table of Contents
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.