Link Check logo

Link Check

Community
jmagly
link-check

Verify @file references in AIWG skills and agents against the linking contract — per-file or corpus-wide, with optional auto-fix

Overview

Publisherjmagly
Repositoryaiwg
Skill namelink-check
Stars
225
Forks
34
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by jmagly on GitHub. Read the source before you install it.

Installation

Install the Link Check AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/jmagly/aiwg/tree/main/agentic/code/addons/aiwg-dev/skills/link-check
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/jmagly/aiwg.git /tmp/aiwg
mkdir -p .claude/skills
cp -r /tmp/aiwg/agentic/code/addons/aiwg-dev/skills/link-check .claude/skills/link-check
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Link Check in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Link Check on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Link Check is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Link Check

You verify every @file reference in AIWG distributable skills and agents against the full linking contract, producing a structured PASS/WARN/FAIL report per reference. You can run against a single file, a directory, or the full agentic/code/ corpus, and optionally auto-migrate legacy bare refs to the @$AIWG_ROOT/ prefix.

Triggers

  • "link check" / "check links" → run against current file or directory
  • "link check agentic/code/" → full corpus scan
  • "link check --fix" → auto-migrate legacy bare refs
  • "any bad links?" → run link-check on current context
  • "check for broken @-refs" → run link-check
  • "are there any legacy bare refs?" → run link-check, focus on 4b

Usage

/link-check [path] [options]
ArgumentDefaultDescription
pathcurrent file or agentic/code/File or directory to check
--fixoffAuto-migrate @agentic/code/, @src/, @docs/, @tools/ → @$AIWG_ROOT/
--reportoffOutput structured JSON report instead of human-readable
--aiwg-onlyoffCheck only .aiwg/ refs (skip bare-core and forbidden checks)
--fail-on-warnoffTreat WARN as FAIL (useful for CI)

Reference Classification

Each @<path> reference is classified per the AIWG linking contract:

PatternResultReason
@$AIWG_ROOT/<path>PASSAIWG core file, install-relative
@$TOKEN/<path> — TOKEN set in envPASSRegistered corpus token
@$TOKEN/<path> — TOKEN NOT setWARNUnknown token — add to .env or export
@.aiwg/<path> — in Tier 1/2 allowlistPASSNormalized project memory
@.aiwg/<path> — not in allowlistFAILRepo-local only — silently fails in user projects
@.claude/<path>FAILDeployment target — forbidden in distributable source
@agentic/code/<path>WARNLegacy bare ref — use @$AIWG_ROOT/agentic/code/
@src/<path>WARNLegacy bare ref — use @$AIWG_ROOT/src/
@docs/<path>WARNLegacy bare ref — use @$AIWG_ROOT/docs/
@tools/<path>WARNLegacy bare ref — use @$AIWG_ROOT/tools/
Relative path within componentPASSLocal ref

Process

  1. Resolve target path — single file, directory, or default agentic/code/
  2. Build normalized allowlist — read all manifest.json files under agentic/code/; collect memory.creates[*].path; add Tier 1 entries
  3. Load environment — check which env vars are set (for @$TOKEN/ classification)
  4. Extract all @<path> refs — regex: @[\$\.]?[a-zA-Z0-9_/.\-]+ in each .md file
  5. Classify each ref per the table above
  6. If --fix: rewrite WARN-classified bare legacy refs in-place:
    • @agentic/code/ → @$AIWG_ROOT/agentic/code/
    • @src/ → @$AIWG_ROOT/src/
    • @docs/ → @$AIWG_ROOT/docs/
    • @tools/ → @$AIWG_ROOT/tools/
  7. Produce report

Output Format

Human-readable (default)

Link Check: agentic/code/addons/aiwg-dev/skills/dev-doctor/SKILL.md

  PASS  $AIWG_ROOT/agentic/code/addons/aiwg-dev/rules/addon-boundaries.md
  PASS  .aiwg/AIWG.md (Tier 1, always present)
  WARN  agentic/code/addons/aiwg-dev/rules/skill-placement.md
        → legacy bare ref; use $AIWG_ROOT/agentic/code/... prefix
  FAIL  .aiwg/planning/issue-driven-ralph-loop-design.md
        → non-normalized .aiwg/ path (repo-local only)

Result: 2 issues (1 WARN, 1 FAIL)

Corpus summary

When run against a directory:

Link Check: agentic/code/ (corpus-wide)

Files checked:  247
References found: 1,341
  PASS:  1,337
  WARN:      3
  FAIL:      1

Violations:
  WARN  agentic/code/addons/my-addon/skills/my-skill/SKILL.md:42
        agentic/code/... (bare ref) → add $AIWG_ROOT/ prefix
  FAIL  agentic/code/addons/my-addon/skills/my-skill/SKILL.md:89
        .aiwg/planning/my-design.md → non-normalized (repo-local only)

Overall: WARN (3 warnings, 1 failure)
Run with --fix to auto-migrate WARN refs.

JSON report (--report)

json
{
  "target": "agentic/code/",
  "timestamp": "2026-04-01T04:48:00Z",
  "summary": { "files": 247, "refs": 1341, "pass": 1337, "warn": 3, "fail": 1 },
  "violations": [
    {
      "file": "agentic/code/addons/my-addon/skills/my-skill/SKILL.md",
      "line": 42,
      "ref": "agentic/code/addons/aiwg-dev/rules/skill-placement.md",
      "result": "WARN",
      "message": "legacy bare ref — add $AIWG_ROOT/ prefix",
      "fix": "$AIWG_ROOT/agentic/code/addons/aiwg-dev/rules/skill-placement.md"
    }
  ]
}

Auto-Fix Behavior (--fix)

Only WARN-classified legacy bare refs are auto-fixed. FAIL refs (repo-local .aiwg/ paths, .claude/ refs) are never auto-fixed — they require a human decision:

  • Remove the reference
  • Replace with a normalized equivalent
  • Or normalize the path in a manifest's memory.creates

After --fix, report how many refs were migrated and re-run the check to confirm the result.

CI Integration

Use --report --fail-on-warn for CI:

bash
# Fails if any FAIL or WARN refs found
aiwg link-check agentic/code/ --report --fail-on-warn > link-report.json

Exit codes:

  • 0 — all PASS
  • 1 — WARN found (only if --fail-on-warn)
  • 2 — FAIL found

Examples

Check a single skill

User: "check links in the dev-doctor skill"

Action: Run link-check on agentic/code/addons/aiwg-dev/skills/dev-doctor/SKILL.md.

Full corpus check

User: "any bad links in the corpus?"

Action: Run link-check on agentic/code/ and summarize violations.

Auto-fix

User: "link check --fix"

Action: Run link-check on agentic/code/, migrate all bare legacy refs, report how many were fixed.

References

  • @$AIWG_ROOT/agentic/code/addons/aiwg-dev/rules/aiwg-dir-reference-contract.md — Full linking contract
  • @$AIWG_ROOT/agentic/code/addons/aiwg-dev/skills/validate-component/SKILL.md — Per-component validation
  • @$AIWG_ROOT/agentic/code/addons/aiwg-dev/skills/dev-doctor/SKILL.md — Full repo health check

Frequently asked questions

What does the Link Check AI skill do?

Verify @file references in AIWG skills and agents against the linking contract — per-file or corpus-wide, with optional auto-fix

Why use Link Check on TypingMind?

Because you install it once and use it with any model. Link Check is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Link Check in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/jmagly/aiwg/tree/main/agentic/code/addons/aiwg-dev/skills/link-check. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Link Check?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Link Check?

As many as you like. As long as a model supports skills, you can use Link Check with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Link Check AI skill free?

Yes. It is published on GitHub by jmagly under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇