These packages give you a linter, validator, and scaffolder for AGENTS.md and SKILL.md files — the markdown documents that define how AI agents behave and what skills they have. You would adopt them to enforce a consistent, machine-readable structure across agent definitions in a multi-agent system, catching formatting errors, missing sections, and broken skill references before they cause runtime issues. The toolkit is built as a pipeline of independent packages (parser → validator → linter → reporter) that share canonical Zod schemas, with an MCP server exposing the same tools directly to AI agents over Stdio or HTTP.
A CLI for linting, validating, scaffolding, and formatting AGENTS.md and SKILL.md files, providing five subcommands (`lint`, `validate`, `scaffold`, `format`, `examples`) with multi-format output (console, JSON, HTML, Markdown) and CI-friendly exit codes. Built on Commander, it also exports utility functions for programmatic use.
A linting engine for AGENTS.md and SKILL.md files that provides 18 built-in rules across style, content, and best-practice categories, exposed as `runLintRules`, `registerRule`, and `runAutoFix` functions. It depends on `@reaatech/agents-markdown-parser` for document parsing and outputs findings in a standard `Finding[]` format.
An MCP server that exposes five tools—`lint_agents_md`, `validate_agents_md`, `validate_skill_md`, `scaffold_agent`, and `get_examples`—for working with agent markdown files via the Model Context Protocol. It exports `createMcpServer()` (returns a configured `Server` instance from `@modelcontextprotocol/sdk`) and `startMcpServer(transportType?)` (starts the server over Stdio or StreamableHTTP).
Pino-based structured logging, OpenTelemetry metrics and tracing, and dashboard aggregation for the `@reaatech/agents-markdown-*` ecosystem, exposed as functions (`info`, `error`, `recordOperation`, `buildDashboardSummary`) and a singleton `MetricsManager` class.
Parses AGENTS.md and SKILL.md markdown files into typed documents with YAML frontmatter extraction, section hierarchy, table parsing, and code block extraction. Exports async functions like `parseMarkdown` and `parseMarkdownFiles` that return `AgentsMdDocument` or `SkillMdDocument` objects, plus utilities for navigating sections, validating frontmatter, and extracting tables and code blocks. Built on remark and unified.
Exports a set of reporter functions (`reportLintResult`, `reportJsonLintResult`, `reportHtmlLintResult`, `reportMarkdownLintResult`, and their validation counterparts) that format lint and validation results into console (ANSI-colorized), JSON, self-contained HTML, or GitHub-flavored Markdown output.
A function that generates AGENTS.md and SKILL.md files from Handlebars templates, creating a complete agent directory structure with one skill file per listed skill ID.
A Zod-based validation engine for AGENTS.md and SKILL.md files that checks frontmatter structure, required sections, section ordering, skill references, MCP tools tables, and content quality against the specification. It exports `validate`, `validateAgentsMd`, and `validateSkillMd` functions that take parsed document objects (from `@reaatech/agents-markdown-parser`) and return structured validation results with errors and warnings.
