Mise en place
for your agents.

Runs the kitchen. Coordinates coding-agent tasks across tools, repos, and sessions with work claims, verification receipts, shared memory, and tool sync. Code intelligence, evidence memory, operator notifications, and publish checks ship inside Brigade.

A skillet full of agent skills. Production-tested workflows for auditing, improving, and shipping repos with AI coding agents: repo audits with prioritized backlogs, bug hunts, security sweeps, publish gates, releases, memory handoffs. Plain SKILL.md files.

Where the chef stores the cookies and the secret recipes. Mirrors the browser sessions and secrets you choose from your daily driver to the machines your agents run on, sealed with AES-256-GCM the whole way.

The bound ledger. MiseLedger is the evidence memory engine built into Brigade: a local evidence graph over agent history, with session, file, git, and chat crawlers, SQLite full-text search, and Markdown export.

Works every station. Now built into Brigade: memory status, lint, and compact keep the Claude Code and OpenClaw memory system healthy. Handoffs go through brigade ingest.

Keeps the mise en place tight. Audits oversize OpenClaw bootstrap files and trims them into reference cards with heuristics plus LLM judgment.

Presents the dish. Agent-neutral desktop capture for people and agents: polished shots with metadata, stable JSON on stdout, an optional MCP server, and vertical video reels. Lifts what is on your screen, frames it, ready to share.

The chalkboard out front. Turns repos, specs, incident reports, and architecture notes into self-contained interactive HTML/SVG explainers: audience modes, source-grounded facts, click-through scenes, exportable artifacts.

Calls the orders. Operator notifications included in Brigade, with push delivery from coding agents to Discord, Telegram, and Signal.

The map of the kitchen. GraphTrail is the code intelligence engine inside Brigade. Its local SQLite graph tracks symbols, imports, and call edges so agents can inspect callers and impact before editing Python, TS/JS, Rust, and Go.

Reduces terminal output before it reaches the agent context. Deterministic compaction for command output.

Finds code locally with semantic and hybrid search, using Ollama and SQLite.

Turns JSON specs and command output into reproducible, sanitized terminal demo SVGs.

Builds local static analytics for agent usage, with API spend and OAuth usage value shown separately.

Supplies the shared theme and OG card template for Escoffier Labs sites, including this page. Proposes version updates for review.