Skip to content

[External Plugin]: WolfMarkDown #2676

Description

@WolfMarkTools

Plugin name

wolfmarkdown

Short description

Creates, cleans, sanitises, formats, and verifies production-ready Markdown using agent judgement plus deterministic validation. Supports safe Markdown composition and repair with integrity, rollback, GFM, markdownlint, Prettier, and idempotence checks.

GitHub repository

WolfMarkTools/WolfMarkDown

Plugin path inside the repository

No response

Ref to review

v0.2.2

Commit SHA to review

6570ec687926bb6b2b37cecd8854970d269d37e5

Version

0.2.2

License identifier

MIT

Author name

WolfMark

Author URL

No response

Homepage URL

No response

Keywords

markdown
agent-skills
formatting
validation
documentation
ai-agents

Additional notes for reviewers

WolfMarkDown is a skills-only Agent Plugins 1.0 package. The repository maintains one canonical skill implementation under wolfmarkdown/; skills/wolfmarkdown is an in-package symlink to ../wolfmarkdown, so marketplace packaging does not create or maintain a second implementation.

The plugin has no MCP server, hosted service, API-key requirement, telemetry, or embedded LLM runtime. Semantic judgement is performed by the host agent; WolfMarkDown's local deterministic tooling handles formatting, linting, GFM parsing, protected-token integrity checks, idempotence, and safe publishing/rollback. Prettier is the sole final Markdown printer.

The deterministic tooling requires Node.js 20+ and uses pinned npm dependencies including Prettier, markdownlint, remark/GFM, unified, and YAML tooling. WolfMarkDown's Setup and Doctor flows handle local runtime/dependency readiness and Agent Skill discovery without requiring an external service.

The submitted immutable v0.2.2 release is backed by 92/92 passing Node tests, skills-ref validate ./wolfmarkdown, and a public semantic-repair reproduction under examples/mobile-app-migration/. That example demonstrates recovery of heading hierarchy, GFM tables and lists from messy AI output while preserving technical values and retaining an explicitly ambiguous source conclusion rather than inventing one.

Public installation through npx skills add WolfMarkTools/WolfMarkDown --skill wolfmarkdown has also been verified, including Agent Skills discovery and a Claude Code-specific global installation.

The immutable v0.2.2 Awesome Copilot submission has passed the complete automated external-plugin intake, including Agent Plugins specification compliance, Vally validation, install smoke testing, version matching, and ref/SHA consistency. Human maintainer review remains pending.

WolfMarkDown deliberately separates agent judgement for semantic structure from deterministic tooling for proof. A WolfMarkDown PASS means the Markdown artifact passed the applicable structural and deterministic checks; it does not claim factual correctness, completeness, currency, policy approval, or authorisation to publish.

Submission checklist

  • The plugin lives in a public GitHub repository.
  • The ref and/or sha I provided is immutable (release tag and/or full 40-character commit SHA), not a branch.
  • This submission follows this repository's contribution, security, and responsible AI policies.
  • This plugin is not already listed in the Awesome Copilot marketplace.

Metadata

Metadata

Assignees

No one assigned

    Labels

    external-pluginPublic external plugin submissionneeds-review:MEDIUMContributor reputation check flagged MEDIUM riskready-for-reviewSubmission passed intake validation and is ready for maintainer review

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions