What output formats are supported?

JSON structured reports for automation and a colored text table for human review.

Which categories can be checked?

Agents, structure, drift, or all of them with the full check.

team-shinchan:lint-harness

Scanned

npx machina-cli add skill seokan-jeong/team-shinchan/lint-harness --openclaw

Files (1)

SKILL.md

1.9 KB

Lint Harness Skill

Check plugin consistency and detect configuration drift.

Usage

/team-shinchan:lint-harness                    # Full check (JSON)
/team-shinchan:lint-harness --category agents   # Agent consistency only
/team-shinchan:lint-harness --category structure # Structural integrity only
/team-shinchan:lint-harness --category drift     # Drift detection only
/team-shinchan:lint-harness --format table       # Colored text table

Arguments

Arg	Default	Description
`--category {name}`	(all)	Check only: `agents`, `structure`, or `drift`
`--format {fmt}`	json	Output format: `json` or `table`

Process

Step 1: Run Lint Script

Execute the harness lint script:

node ${CLAUDE_PLUGIN_ROOT}/src/harness-lint.js [args]

Step 2: Display Results

Show the output from the lint script directly to the user.

JSON format: structured report with categories, checks, pass/fail, details
Table format: colored text table with PASS/FAIL indicators

Step 3: Summarize

After displaying results, provide a brief summary:

Harness Lint: {passed}/{total} checks passed
{list any FAIL items with recommended fixes}

Category	What it checks
agents	Required frontmatter, coding-principles refs, maxTurns, permissionMode
structure	Skill-command parity, hook registration, cross-refs, layer-map
drift	Output-formats refs, self-check refs, version consistency

Important

Script writes .shinchan-docs/.last-lint timestamp after running
Recommended to run at least once per week
Zero external dependencies — uses only Node.js built-in modules

Source

git clone https://github.com/seokan-jeong/team-shinchan/blob/main/skills/lint-harness/SKILL.mdView on GitHub

Overview

Lint Harness checks plugin consistency and detects configuration drift across Shinchan skills. It validates agent frontmatter, structural integrity, and drift, and can emit output as JSON for automation or a colorized table for humans. No external dependencies are required beyond Node.js.

How This Skill Works

Run the harness-lint script via Node to perform the selected categories (agents, structure, drift) or all categories. The tool outputs a structured JSON report or a table, followed by a brief PASS/FAIL summary with details on any failing checks.

When to Use It

During CI to validate plugin changes before merging
On a weekly basis to catch configuration drift
When adding or updating agent frontmatter or required fields
After making structural changes (parity, hooks, cross-refs) to ensure integrity
Before a release to confirm overall consistency across categories

Quick Start

Step 1: Run the lint script: node ${CLAUDE_PLUGIN_ROOT}/src/harness-lint.js [args]
Step 2: Choose format and categories (e.g., --category agents --format json)
Step 3: Review the JSON/table output and the summary to fix any issues

Best Practices

Run the full check regularly and after any plugin update
Use --format json in automation pipelines to parse results
Always verify all three categories: agents, structure, drift
Review the summary for any FAIL items and apply fixes promptly
Keep a local history by noting the .shinchan-docs/.last-lint timestamp

Example Use Cases

CI job outputs a JSON report showing PASS/FAIL per category for a new plugin
Detects missing required agent frontmatter in a recently added skill
Identifies a mismatch between skill-command parity and hook registrations
Flags drift in output-formats references after a plugin upgrade
Verifies version consistency across self-check refs before deployment

Frequently Asked Questions

Add this skill to your agents