Claude Skills by emlab-ai

How an agent works with Wye through the wye CLI — resolve a Wye link (http://…/<product>/<project>/d/<doc>#…) or node id to its text, read and write documents and nodes, search product knowledge, and report progress on a session. Use whenever a message, task or prompt contains a Wye link, a node id like req:x / task:y / goal:z, a session id, or asks to update the PRD, plan, tech design or knowledge of a product kept in Wye.

The read/write contract between a coding agent and the product context graph (docs/context-graph). Use BEFORE implementing or changing any feature in a repo that has a context graph — query the graph with wye instead of exploring the codebase, describe the change as a delta before coding, and run wye check before claiming done. Also use when asked "what does the system say about X", "what depends on Y", or "is Z already a requirement".

Build the product context graph for one module — requirements, entities, state machines, ops, pages/actions, rules with sources, gates, tests, drift — as data/products/<product>/projects/<project>/docs/<module>.md, then build/check it with wye and publish the phone viewer. Use when asked to "describe module X", "graph the Y feature", "add Z to the context graph", or to repeat the inventory pilot for another area.

Restore and continue a Wye agent session by id — /wye-restore <session-id> [--product <p>]. Loads the session's instruction, referenced context, log and result, then continues the work from where the previous agent stopped, logging progress back to the same session.

_What a session does when it runs this skill — the instruction, in markdown. Tag the knowledge it needs (`kind:slug`); `takes:` and `writes:` on the card say which kind it runs on and which it produces._