github-triage

Read-only GitHub triage for issues AND PRs. 1 item = 1 background task (category: quick). Analyzes all open items and writes evidence-backed reports to /tmp/{datetime}/. Every claim requires a GitHub permalink as proof. NEVER takes any action on GitHub - no comments, no merges, no closes, no labels. Reports only. Triggers: 'triage', 'triage issues', 'triage PRs', 'github triage'.

Quality

75%

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 ./.opencode/skills/github-triage/SKILL.md

Quality

Content

57%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 highly actionable and well-phased, but it suffers from repeated policy restatements, an unreferenced bundle fetcher script with duplicated inline logic, and missing validation checkpoints for a batch operation. Structure exists but content that should live in separate files is inlined.

Suggestions

Reference scripts/gh_fetch.py for fetching/pagination instead of inlining ~40 lines of bash, so the bundle file is actually used and SKILL.md stays lean.

Dedupe the zero-action policy to a single authoritative section and have subagent prompts point back to it rather than restating forbidden commands and footers repeatedly.

Add an explicit validation checkpoint in Phase 4 (verify each report file exists, contains at least one permalink, and that no gh mutation command was issued) before marking a task completed.

Dimension	Reasoning	Score
Conciseness	Mostly concrete with no basic-concept padding, but the zero-action policy is restated 4+ times (role, <zero_action>, ABSOLUTE RULES, Common Preamble, and each PR prompt's footer) and the forbidden-commands list is duplicated, so it could be tightened substantially.	3 / 5
Actionability	Provides copy-paste-ready bash for fetching/pagination and complete per-type report templates, plus concrete gh commands; minor gaps come from the pseudocode task_create/task/background_output API that isn't a real tool signature.	4 / 5
Workflow Clarity	A clear Phase 0–5 sequence exists, but this is a batch operation (spawn one task per item, process ALL items) and Phase 4 lacks validation that reports were written, contain evidence, or that no GitHub mutation occurred — the batch-operation cap of 3 applies.	3 / 5
Progressive Disclosure	Section structure is present, but the bundle file scripts/gh_fetch.py is never referenced while ~40 lines of pagination bash are inlined instead, and ~300 lines of per-type subagent prompt templates that belong in reference files are all inlined in SKILL.md.	3 / 5
	Total	13 / 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, specific description in third-person voice that clearly states capabilities, boundaries, and concrete trigger phrases. Minor room for richer trigger-term synonym coverage, but it answers what and when unambiguously.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'Analyzes all open items', 'writes evidence-backed reports to /tmp/{datetime}/', 'Every claim requires a GitHub permalink', 'NEVER takes any action on GitHub' — covering the skill comprehensively.	5 / 5
Completeness	Explicitly answers both 'what' (read-only triage for issues AND PRs, evidence-backed reports, no mutations) and 'when' via a concrete 'Triggers:' clause with specific phrases.	5 / 5
Trigger Term Quality	Includes natural trigger phrases users would say ('triage', 'triage issues', 'triage PRs', 'github triage') with some variation, but synonyms are limited and there are no file-extension-style cues; a few natural phrasings are missing.	4 / 5
Distinctiveness Conflict Risk	Clear niche — read-only GitHub triage with permalink-evidence requirement and explicit zero-mutation boundary — with distinct triggers, making overlap with other skills minimal.	5 / 5
	Total	19 / 20 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
skill_md_line_count	SKILL.md is long (588 lines); consider splitting into references/ and linking	Warning
relative_links	Relative link issues: 1 missing	Warning

	Total	14 / 16 Passed

Repository: code-yeongyu/oh-my-openagent
Path: .opencode/skills/github-triage/SKILL.md
Commit: 038ed0c

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.