16 machine-discoverable capabilities
Agents can discover capabilities by problem, inspect requirements and boundaries, and consume the same records as JSON or Markdown without scraping presentation markup.
Capabilities
Each capability publishes the problem, outcomes, requirements and boundaries an agent needs before choosing it.
Diagnose and repair external-write failures without duplicate side effects by ranking evidence, separating ambiguous outcomes, and requiring authoritative provider readback before success.
Build a resumable coding-agent workflow that selects bounded work from repository evidence, persists decisions in a durable ledger, verifies work before completion, and stops at explicit approval boundaries.
Map a live software system without mutating source, services, schedules, databases, credentials, providers, or runtime state, while reconciling documentation against observed runtime truth.
Build an evidence-bound execution system where one bounded consequential action requires transaction-specific approval, executes exactly once, is independently verified, and produces an offline-verifiable signed receipt.
Problem routing
Category and problem signals come directly from the canonical catalogue.
Machine discovery
JSON, text and Markdown projections derive from the same canonical records as the HTML routes.
Canonical public catalogue projection for deterministic agent discovery.
Site-specific agent discovery map: routes, inventory, policy and future CLI boundary.
Convenience index for language-model consumers; not claimed as a universal protocol.
Per-capability metadata: problem, outcomes, requirements, boundaries and source pointer.