harness-setup

HAR: Project init, tool setup, agent config, memory setup, skill mirror sync. Trigger: setup, init, new project, CI/Codex setup, harness-mem, mirror. Do NOT load for: implementation, review, release, planning.

Quality

79%

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

Fix and improve this skill with Tessl

tessl review fix ./opencode/skills/harness-setup/SKILL.md

Quality

Content

71%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-structured dispatch hub that routes subcommands to real reference files with concrete verification commands and a clear execution sequence. The main weakness is redundancy between the Quick Reference table and the Reference Index, plus dense policy-anchor blockquotes that pad the body.

Suggestions

Consolidate the Quick Reference table and the Reference Index into a single section to remove the duplicated file listing and save tokens.

Move the dense 'Claude Code setup guidance' and 'Codex plugin workflows' policy blockquotes into the relevant reference files (codex.md, ci.md), keeping only a one-line pointer in SKILL.md.

Inline the key dry-run and verification command arguments in the Execution step 4 (e.g., the expected `harness doctor` exit behavior) so verification does not require opening a reference file first.

Dimension	Reasoning	Score
Conciseness	Mostly efficient hub content, but the Quick Reference table and the Reference Index repeat the same file list, and the multi-paragraph policy-anchor blockquotes add padding that could be tightened or moved to references.	3 / 5
Actionability	Provides concrete dispatch commands ('/harness-setup init'), verification commands ('bash scripts/sync-skill-mirrors.sh --check', 'bash scripts/ci/check-consistency.sh', 'harness doctor'), and a 4-step Execution sequence; minor gaps where full command arguments live only in the referenced files.	4 / 5
Workflow Clarity	The Execution section is a clear 4-step sequence with dry-run-then-execute guidance (step 3) and an explicit verification step (step 4) calling doctor and check scripts; checkpoints are present but partly delegated to the reference files rather than fully inline.	4 / 5
Progressive Disclosure	Clear overview with well-signaled one-level-deep references via '${CLAUDE_SKILL_DIR}/references/*.md'; all eight referenced files exist, navigation is organized in both a Quick Reference table and Reference Index, and there are no nested reference chains.	5 / 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, trigger-rich description that clearly states both what the skill does and when to load it, plus a negative trigger list to avoid misfires. Specificity is good though the action phrasing is terse.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ('Project init, tool setup, agent config, memory setup, skill mirror sync') but the action labels are terse domain nouns rather than fully spelled-out operations, leaving minor coverage gaps.	4 / 5
Completeness	Explicitly answers what ('Project init, tool setup, agent config, memory setup, skill mirror sync') and when via the 'Trigger:' clause plus a 'Do NOT load for' exclusion list, matching the top anchor.	5 / 5
Trigger Term Quality	'Trigger: setup, init, new project, CI/Codex setup, harness-mem, mirror' gives good natural keyword coverage a user would actually say, though a few common synonyms are absent.	4 / 5
Distinctiveness Conflict Risk	The harness-specific trigger niche combined with the explicit 'Do NOT load for: implementation, review, release, planning' exclusion gives a clear niche with minimal conflict risk.	5 / 5
	Total	18 / 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
referenced_paths_exist	Referenced path issues: 2 missing, 1 deeper-than-1-level	Warning

	Total	15 / 16 Passed

Repository: Chachamaru127/claude-code-harness
Path: opencode/skills/harness-setup/SKILL.md
Commit: 7d8deba

Reviewed: about 15 hours 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.