self-improvement

Appends new entries to LESSONS-LEARNED.md via opencastle lesson CLI; searches past lessons for matching errors; proposes skill updates when retry patterns exceed thresholds. Use when consulting or updating LESSONS-LEARNED.md, after task failures, when capturing retrospective insights, or when a retry succeeds.

Quality

85%

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

78%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.

A concise, largely actionable instruction skill with a well-sequenced validated workflow. Its main weaknesses are a dangling reference to a missing LESSON-CATEGORIES.md (which also hides required CLI values) and an undefined retry threshold mentioned in both workflow and description.

Suggestions

Add LESSON-CATEGORIES.md to the bundle (or inline the valid --category/--severity values in the body) so the referenced taxonomy and required CLI values are actually reachable.

Replace 'up to threshold' in step 3 with a concrete number of retries (e.g., 'up to 3 retries'), since the threshold is referenced but never defined.

Make the verification step state what a well-formed entry looks like (e.g., a sample appended line) so 'If malformed' is judged against a concrete example rather than implied.

Dimension	Reasoning	Score
Conciseness	Lean and efficient — ~53 lines, no explanations of concepts Claude already knows, every section (core rule, CLI command, workflow, quality rules, anti-patterns) earns its tokens.	5 / 5
Actionability	The 'opencastle lesson' command with all flags and the concrete 'rg' search example are copy-paste ready, but valid --category/--severity values are deferred to LESSON-CATEGORIES.md which is not in the bundle, so a first run could fail on an invalid value.	4 / 5
Workflow Clarity	Clear 6-step sequence with an explicit validation step (verify via 'tail -1', re-run if malformed) and feedback loops, but step 3's 'up to threshold' references a retry threshold that is never defined anywhere in the skill.	4 / 5
Progressive Disclosure	Sections are well-organized and the body is near the simple-skill size, but the sole external reference ([LESSON-CATEGORIES.md](LESSON-CATEGORIES.md)) points to a file that does not exist in the bundle — the needed category/severity taxonomy is unavailable.	3 / 5
	Total	16 / 20 Passed

Description

92%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.

A strong description: concrete third-person action verbs, a distinct niche named via specific file and CLI, and an explicit 'Use when...' clause with multiple natural trigger phrases. Only minor keyword synonyms are missing.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete third-person actions — 'Appends new entries to LESSONS-LEARNED.md via opencastle lesson CLI', 'searches past lessons for matching errors', 'proposes skill updates' — with comprehensive coverage of the skill's capabilities.	5 / 5
Completeness	Explicitly answers both what (three concrete actions) and when ('Use when consulting or updating LESSONS-LEARNED.md, after task failures, when capturing retrospective insights, or when a retry succeeds') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Good natural triggers ('after task failures', 'when capturing retrospective insights', 'when a retry succeeds', 'consulting or updating LESSONS-LEARNED.md') but misses common synonyms users might say such as 'post-mortem', 'mistake', or 'what went wrong'.	4 / 5
Distinctiveness Conflict Risk	Clear niche anchored to a specific file (.opencastle/LESSONS-LEARNED.md) and specific CLI (opencastle lesson), with triggers unlikely to fire for unrelated skills; uses third person voice throughout.	5 / 5
	Total	19 / 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
relative_links	Relative link issues: 1 missing	Warning

	Total	15 / 16 Passed

Repository: monkilabs/opencastle
Path: src/orchestrator/skills/self-improvement/SKILL.md
Commit: c0f4566

Reviewed: 1 day 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.