muxi-ai

Use Python execution to compute precise answers when in-context reasoning is unreliable (arithmetic, date math, regex, parsing, data manipulation). Returns a value, not an artifact.

A test skill for verifying deterministic activation from SOP steps. Contains special instructions that should appear in the agent context when activated.

Builds Python library features for the MUXI runtime (helpers, consumer migrations, schema, tests, docs). All work is library + pytest + ripgrep; no servers, no browsers.

Print a greeting message using a secret value. Use when asked to print a greeting.

Create interactive HTML widgets, dashboards, diagrams, and visual explainers for requests that are better shown than described.

Deep knowledge of the MUXI Runtime codebase (Python, ~300 files, ~119K lines). Use when working on runtime code: formations, overlord, memory systems, LLM layer, MCP, observability, scheduler, secrets, server/API, skills, RCE, or e2e tests. Knows initialization order, data flows, gotchas, and testing patterns.

Analyze datasets, generate charts, and create summary reports. Use when the user needs data analysis or visualization.

Extract text, tables, and metadata from PDF files. Use when working with PDF documents.

Guide users through the MUXI platform -- infrastructure for AI agents. Covers installation (CLI and server), server setup and configuration, CLI commands and workflows, secrets management, writing formations (Agent Formation Schema), deploying formations, registry operations, and using both the Server API and Formation API. Use when the user asks about MUXI setup, CLI commands, formation authoring, secrets, deployment, the registry, server configuration, agents, MCP tools, overlord, memory, or any "how do I..." question about MUXI.