CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

github-bot

Interact with the GitHub API as the joelclawgithub[bot] app. Use when creating PRs, pushing commits, managing issues, checking CI status, listing repos, or any GitHub API operation. Triggers on "create a PR", "push to GitHub", "open an issue", "check CI", "list repos", "merge PR", "GitHub API", or any task requiring programmatic GitHub access. All actions appear as joelclawgithub[bot], not as @joelhooks.

Quality

86%

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

SKILL.md
Quality
Evals
Security

Quality

Content

68%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 compact, largely executable reference for authenticating and calling the GitHub API as the bot app, with a real helper script and a sensible 401-retry loop. Its main weaknesses are validation gaps around the destructive actions it advertises (push, merge, PR creation) and a couple of off-topic sections (ShitRat preamble, Receipts) that add confusion without adding instruction.

Suggestions

Add verification checkpoints for the destructive actions the skill lists (create-pr, push-commit, merge-pr): e.g., after creating a PR, capture and report the PR URL from the API response; after pushing, verify the ref via a follow-up GET before declaring success.

Trim or move the tangential content: the ShitRat preamble and the Receipts section (OTel / Brain .svx) either belong in a short 'Scope / deprecated patterns' note or should be cut, since they don't instruct the reader how to operate the GitHub API.

Replace placeholder endpoints ('repos/OWNER/REPO/...', 'repos/endpoint') with one concrete end-to-end example (e.g., creating a PR on a named repo with the token) so the curl pattern is copy-paste ready, and consider moving the pagination loop into scripts/ as a helper.

DimensionReasoningScore

Conciseness

Largely lean and operational (token command, curl pattern, secrets table, pagination loop), but the opening ShitRat paragraph ("Discover actual app access; old repository counts are not authority") and the Receipts section's OTel / Brain ".svx" tangents are trimmable noise that don't serve the API task.

4 / 5

Actionability

Mostly executable: a real token-generation script (scripts/github-token.sh exists), a concrete curl header pattern, and a working pagination bash loop. Minor gaps — "repos/OWNER/REPO/..." and "repos/endpoint" are placeholder pseudocode rather than a fully copy-paste-ready example, and SKILL_DIR must be substituted.

4 / 5

Workflow Clarity

The auth workflow has a good feedback loop ("Generate a new one if a request returns 401"), but the skill explicitly covers destructive operations — "push-commit", "merge-pr", "create-pr" — with no verification steps (e.g., confirm PR URL, check merge status, verify CI after push), so validation gaps cap this at 3 per the destructive-operations guideline.

3 / 5

Progressive Disclosure

Well-organized sections (Authentication, Secrets, Bot Identity, Pagination) and a correctly-resolved one-level-deep reference to scripts/github-token.sh, which exists. Slightly above the under-50-line simple-skill threshold, and the pagination loop and secrets list are candidates for scripts/references files; good but not perfectly split.

4 / 5

Total

15

/

20

Passed

Description

96%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 capabilities, natural trigger phrases, explicit what/when guidance, and identity disambiguation. It sits between anchors 4 and 5 on distinctiveness due to the broad catch-all triggers, but otherwise matches the top anchors cleanly.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — "creating PRs, pushing commits, managing issues, checking CI status, listing repos" — plus the bot identity behavior, giving comprehensive coverage of GitHub API operations with no vague filler.

5 / 5

Completeness

Explicitly answers both "what" ("Interact with the GitHub API as the joelclawgithub[bot] app") and "when" ("Use when creating PRs..." plus "Triggers on...") with concrete trigger phrases, matching the top anchor example.

5 / 5

Trigger Term Quality

Natural trigger phrases users would actually say: "create a PR", "push to GitHub", "open an issue", "check CI", "list repos", "merge PR", "GitHub API" — covering synonyms and variations comprehensively.

5 / 5

Distinctiveness Conflict Risk

Mostly distinct: bot-app identity, specific trigger phrases, and the "not as @joelhooks" disambiguator. Minor overlap risk remains with any general GitHub/gh-CLI skill because of the catch-all "any GitHub API operation" / "any task requiring programmatic GitHub access" clauses.

4 / 5

Total

19

/

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

joelhooks/joelclaw

Path

skills/github-bot/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.