self-improvement-ci

CI-only self-improvement workflow using gh-aw (GitHub Agentic Workflows). Captures recurring failure patterns and quality signals from pull request checks, emits structured learning candidates, and proposes durable prevention rules without interactive prompts. Use when: you want automated learning capture in CI/headless pipelines.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/self-improvement-ci/SKILL.md

Quality

Content

75%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 well-organized, actionable skill body with concrete commands, a clear output schema, and an appropriately separated reference file. It could tighten the Context Limitation section and add explicit verification of the emitted output against the schema.

Suggestions

Add an explicit validation step that confirms emitted candidates conform to the Output Schema before finalizing a run.

Tighten the "Context Limitation" section by collapsing the implications bullets into a single concise statement.

Make the Heal Handoff Intake steps more concrete with example commands or a code snippet for reading .learnings/HEALS.md.

Dimension	Reasoning	Score
Conciseness	The body is largely lean with concrete commands and a compact output schema, avoiding explanations of concepts Claude already knows; the "Context Limitation" implications list is slightly verbose and could be tightened.	4 / 5
Actionability	It provides executable commands (gh skill install, gh aw compile --validate --strict, gh aw run) and a complete YAML output schema; the Heal Handoff Intake is step-by-step but descriptive rather than copy-paste code, leaving minor gaps.	4 / 5
Workflow Clarity	The Authoring Workflow and Heal Handoff Intake are clearly numbered sequences with an explicit validation checkpoint ("Validate: gh aw compile --validate --strict"); the read-only CI contract means the destructive/batch cap does not apply, but a couple of checkpoints are implicit.	4 / 5
Progressive Disclosure	The body is a well-sectioned overview with a single clearly-signaled one-level-deep reference ("Example-only templates live in references/workflow-example.md"), and the referenced file exists and matches; minor organization gaps keep it below 5.	4 / 5
	Total	16 / 20 Passed

Description

83%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, specific description with an explicit 'Use when' trigger and concrete capability list. The main weakness is second-person voice in the trigger clause and slightly narrow trigger-term variety.

Suggestions

Rewrite the trigger clause in third person to avoid the second-person penalty, e.g. "Use when automated learning capture is needed in CI or headless pipelines".

Broaden trigger terms with synonyms users might actually say, such as "CI learning loop", "headless self-improvement", or "PR check learnings".

Dimension	Reasoning	Score
Specificity	"Captures recurring failure patterns and quality signals from pull request checks, emits structured learning candidates, and proposes durable prevention rules" lists three concrete, comprehensive actions; reduced from 5 to 4 because the trigger clause uses second person ("you want"), which the rubric penalizes by one point.	4 / 5
Completeness	It explicitly answers both what (captures patterns, emits candidates, proposes rules) and when ("Use when: you want automated learning capture in CI/headless pipelines") with a concrete trigger phrase.	5 / 5
Trigger Term Quality	Terms like "CI/headless pipelines", "pull request checks", and "automated learning capture" are natural for this niche and cover the main trigger surface, though a few common synonyms or variants are missing.	4 / 5
Distinctiveness Conflict Risk	The "CI-only" scope, gh-aw tooling, and PR-check focus carve a clear niche distinct from the interactive "self-improvement" skill, with only minor overlap risk against that sibling.	4 / 5
	Total	17 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: pskoett/pskoett-ai-skills
Path: skills/self-improvement-ci/SKILL.md
Commit: 8a71d70

Reviewed: 11 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.