common-learning-log

Append a learning entry to AGENTS_LEARNING.md when an AI agent makes a mistake. Auto-activates after a pre-write audit auto-fix, a retrospective correction loop, or a mid-session user correction. Use when: mistake, wrong, correction, my bad, agent error, learning log.

Quality

81%

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

72%Weight 40%Scale 1-3

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

The content is well-structured with concrete, actionable steps and clean progressive disclosure to a single real reference file. The main gaps are a padded, partly malformed "Canonical response anchors" section and the absence of any verification checkpoint in the workflow.

Suggestions

Remove or fix the "Canonical response anchors" section — entries like "Append to AGENTSLEARNING,append" appear malformed and add tokens without actionable value.

Add a brief verification checkpoint to the Protocol (e.g., confirm the new Iteration header was appended and N+1 is correct) to lift workflow clarity.

Tighten Guidelines lines that read as fragments ("that wrong", "state what to , not what to avoid") which contain missing words.

Dimension	Reasoning	Score
Conciseness	The body is lean and avoids explaining concepts Claude already knows, but the "Canonical response anchors" section ("Append to AGENTSLEARNING,append", "Pre-write; trigger") is low-value padding with a malformed entry that could be trimmed. It is not 3 because unnecessary content is present, and not 1 because most of the body is efficient.	2 / 3
Actionability	Concrete, executable guidance is given: "count existing `## Agent Learning Log: Iteration` headers → N" and "write Iteration #(N+1) using format in [Log Entry Format]", backed by a real template file. It is not 2 because exact header patterns and a copy-paste-ready reference are provided rather than vague or pseudocode direction.	3 / 3
Workflow Clarity	The four-step Protocol (Detect signal → Read/count → Append → Continue) is clearly sequenced and numbered, but there is no verification checkpoint confirming the entry was written or that N was counted correctly. It is not 3 because no explicit validation step exists, and not 1 because the sequence is clear and ordered.	2 / 3
Progressive Disclosure	The overview body points to a real, one-level-deep reference (references/log-format.md) via inline links and a References section, with template/detail content appropriately split out. It is not 2 because references are well-signaled and genuinely one level deep, and not 1 because there is no nested chain.	3 / 3
	Total	10 / 12 Passed

Description

90%Weight 40%Scale 1-3

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 clearly communicates a single concrete action and an explicit, natural-language trigger set with good distinctiveness. Its only weakness is that it names one primary action rather than a broader set of concrete capabilities.

Dimension	Reasoning	Score
Specificity	"Append a learning entry to AGENTS_LEARNING.md" names one concrete primary action with surrounding trigger context, but it does not enumerate multiple distinct concrete actions like the level-3 anchor. It is not 1 because the action is concrete rather than vague, and not 3 because only a single main action is described.	2 / 3
Completeness	It states what ("Append a learning entry to AGENTS_LEARNING.md") and when (explicit "Use when:" clause plus auto-activation conditions). It is not 2 because the when is explicitly stated rather than merely implied.	3 / 3
Trigger Term Quality	"Use when: mistake, wrong, correction, my bad, agent error, learning log" covers natural correction language a user would actually say. It is not 2 because coverage is broad and natural rather than missing common variations.	3 / 3
Distinctiveness Conflict Risk	The niche (AI mistake learning log tied to AGENTS_LEARNING.md with correction-derived triggers) is distinct and unlikely to fire for unrelated skills. It is not 2 because the specific file and correction context reduce overlap with similar skills.	3 / 3
	Total	11 / 12 Passed

Validation

87%

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 — 14 / 16 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	14 / 16 Passed

Repository: HoangNguyen0403/agent-skills-standard
Path: .github/skills/common/common-learning-log/SKILL.md
Commit: 9f695e8

Reviewed: 6 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.