CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

learning-aggregator-ci

[Beta] CI-only learning aggregation workflow using gh-aw (GitHub Agentic Workflows). Scans .learnings/ files on a schedule, groups entries by pattern_key, identifies promotion-ready patterns, and posts a gap report as a PR or issue comment. Use when: you want automated cross-session pattern detection in CI/headless pipelines without interactive prompts. For interactive use, use learning-aggregator.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

The canonical home for this skill is learning-aggregator-ci in pskoett/pskoett-ai-skills

SKILL.md
Quality
Evals
Security

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.

The body is dense and highly actionable, with copy-paste install commands, a complete YAML output schema, and a real one-level-deep workflow template in references/. Its main weakness is redundancy: the Workflow Rules, trigger YAML, and differences table restate content that already lives in the reference file or in earlier sections, which hurts token efficiency and blurs the overview-to-detail split.

Suggestions

Cut the duplicated `on:` trigger YAML from the Trigger Configuration section and point readers to the frontmatter of references/workflow-example.md instead — the reference already contains the identical block (conciseness).

Compress the 12-step Workflow Rules into a short list of invariants (read-only, exact-match grouping, promotion threshold, single consolidated comment) and defer the step-by-step detail to references/workflow-example.md, which already contains the full 18-step version (conciseness, progressive_disclosure).

Add error-recovery guidance to the Authoring Workflow — e.g., what to check and fix when `gh aw compile` or `--actionlint --zizmor` fails — to close the feedback-loop gap (workflow_clarity).

DimensionReasoningScore

Conciseness

The prose is dense and free of fluff, but there is real redundancy: the 12-step "Workflow Rules" section restates the detailed steps already in references/workflow-example.md, the Trigger Configuration `on:` YAML duplicates the reference template's frontmatter, and the "Differences from Interactive Version" table repeats the Purpose and Context Limitation sections. This fits anchor 3's "mostly efficient but could be tightened" better than anchor 4's "minor instances".

3 / 5

Actionability

Copy-paste-ready install commands ("gh skill install pskoett/pskoett-skills learning-aggregator-ci", "npx skills add ..."), a concrete validation command ("gh aw compile --actionlint --zizmor"), and a complete output-schema YAML with realistic values cover the common setup and execution cases, matching anchor 5.

5 / 5

Workflow Clarity

The authoring workflow is clearly sequenced with an explicit validation checkpoint (copy → customize → validate → push) and the 12 execution rules are ordered, but anchor 5's "feedback loops for error recovery" are missing — there is no guidance on what to do when `gh aw compile` fails or when comment posting fails.

4 / 5

Progressive Disclosure

The bundle reference references/workflow-example.md is real, well-signaled, and exactly one level deep, and sections are clearly organized. However, the detailed agent rules are inlined in the body rather than deferred to the reference, so content is not cleanly split between overview and detail (anchor 4's "minor organization gaps").

4 / 5

Total

16

/

20

Passed

Description

87%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: it concretely states what the skill does and when to reach for it, with explicit CI/headless trigger phrasing and a clean redirect to the interactive variant. The only deductions are the second-person "you want" phrasing in the Use-when clause and a few missing natural trigger synonyms.

DimensionReasoningScore

Specificity

"Scans .learnings/ files on a schedule, groups entries by pattern_key, identifies promotion-ready patterns, and posts a gap report as a PR or issue comment" lists four concrete, comprehensive actions (anchor 5), but the second-person "Use when: you want automated cross-session pattern detection" triggers the mandated 1-point specificity penalty, capping the score at 4.

4 / 5

Completeness

It explicitly answers both "what" (scans, groups, identifies promotion-ready patterns, posts a gap report) and "when" ("Use when: you want automated cross-session pattern detection in CI/headless pipelines without interactive prompts") with concrete trigger phrases, matching the anchor 5 example structure.

5 / 5

Trigger Term Quality

Natural phrases like "CI-only", "on a schedule", "cross-session pattern detection", "CI/headless pipelines", and "gap report" give good keyword coverage, but anchor 5's comprehensive synonym coverage is not met (e.g., no "GitHub Actions", "scheduled report", or similar variations a user might say).

4 / 5

Distinctiveness Conflict Risk

"CI-only learning aggregation workflow using gh-aw" carves a clear niche, and "For interactive use, use learning-aggregator" explicitly disambiguates the closest sibling skill, so conflict risk is minimal.

5 / 5

Total

18

/

20

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

pskoett/pskoett-ai-skills

Path

plugin/skills/learning-aggregator-ci/SKILL.md

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.