mycelium

Run end-to-end smoke tests for the Mycelium stack. Verifies install, memory, search, coordination, and OpenClaw integration. Use when validating a release, after a deploy, or when something feels broken.

A/B test multi-agent consensus quality with and without Mycelium's structured negotiation, using fully-composed agent personas from the agent-personas dataset. Pick a named scenario (e.g. ex07_investment_portfolio) and the skill fetches the right personas, builds SOUL.md for each agent, then runs the standard before/after flow.

Use the mycelium CLI to join coordination rooms, negotiate with other agents via CognitiveEngine, and share persistent memory across sessions.

Multi-agent coordination layer with persistent memory. Use when coordinating with other agents, sharing context across sessions, joining coordination rooms, or searching shared knowledge. Triggers on "coordinate", "negotiate", "share memory", "session join", "mycelium", "what do other agents think".

Use the mycelium CLI to join coordination rooms, negotiate with other agents via CognitiveEngine, and share persistent memory across sessions.

Run end-to-end smoke tests for the Mycelium claude_code adapter. Verifies claude CLI prereqs, daemon installation, single-host cold-spawn dispatch, multi-room ownership semantics, notes persistence across spawns, control verbs (status/abort), budget gating, concurrent dispatch serialization, and (with funded API credits) autonomous coordination via mycelium-daemon coordination_tick handling. Use when validating the claude_code adapter after changes to `integrations/claude_code/**`, the daemon, or after upgrading the `claude` CLI itself.

Run end-to-end smoke tests for the Mycelium cursor adapter. Verifies cursor-agent prereqs, single-host dispatch, multi-host dispatch through the hub, cross-family negotiation with claude_code/openclaw, workspace asset drift, and auth failures. Use when validating the cursor integration on a fresh install, after touching cursor-family code (`integrations/cursor/**`, `daemon/dispatch.py`, `daemon/runner.py`), or after upgrading `cursor-agent` itself.

Diagnose and fix common Mycelium installation and runtime issues. Use when encountering errors with mycelium commands, backend connectivity, Docker containers, LLM configuration, memory operations, or database migrations. Triggers on "not working", "error", "failed", "cannot connect", "troubleshoot", "debug", "fix".

Cut a release — commit any staged changes, tag, create GitHub release, and optionally notify Webex. Use when the user says /release, /release --with-webex, or /release --preview.

Run precommit checks (lint, format, tests) on the mycelium codebase. Use when the user says /precommit or wants to check code quality before committing.

Regenerate the HTML docs site (CLI Reference, Configuration, content sections) from decorators, the pydantic config schema, and markdown source files

Ship changes — run precommit checks, commit, and optionally push. Use when the user says /ship or wants to commit and push their work.

Modern Python tooling practices for the mycelium project. Use when installing dependencies, running tools, managing environments, or writing Python code. Always active as a background reference.

Test-driven development workflow. Use when the user says /tdd or wants to build a feature test-first. Guides red-green-refactor cycle.