Skip to content

Skills

Plugin

A skill is a slash-command-backed unit of expertise that the agent loads on demand. Each one carries focused knowledge — when to use a document type, required sections, typical relations, common mistakes — so you describe intent instead of schema.

The plugin ships 32 skills in three families:

FamilyCountWhat it doesInvoke via
Intent commands8Route a fuzzy intent to the right document type or workflow/archcore:capture, /archcore:plan, …
Document-type skills18Create and link one document type with the right template/archcore:adr, /archcore:prd, …
Tracks6Orchestrate multi-document flows with automatic relation wiring/archcore:product-track, /archcore:iso-track, …

Plus one utility: /archcore:verify for plugin integrity checks.

  1. You type a slash command (/archcore:adr) or describe an intent (“record a decision about caching”).
  2. The host matches the request to a skill definition in the plugin package.
  3. The skill is loaded into the conversation — its description, guardrails, and relation rules become part of the agent’s working context.
  4. The agent executes the skill, calling MCP tools as needed.

Every skill ultimately resolves to MCP tool calls. The plugin never writes to .archcore/ directly.

  • Start with an intent command when you’re not sure which document type fits. /archcore:capture and /archcore:decide will pick for you.
  • Jump to a document-type skill when you know what you’re creating. /archcore:adr, /archcore:rule, /archcore:spec.
  • Use a track when one document alone won’t do the job. Need a decision and a rule and a how-to guide? Use /archcore:standard-track.

See the dedicated pages for each family:

For a single-page summary of everything, see the Plugin Skills Reference.

Inside any plugin-enabled session:

/archcore:help

Returns the list of available skills, grouped by family, with one-line descriptions. Useful when you forget the exact command name.