Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.qredence.ai/llms.txt

Use this file to discover all available pages before exploring further.

Auto-generated overview of the monorepo workspace. Start with the Introduction and Quickstart if you are new to Fleet Pi.

​
Workspace layout

fleet-pi/
├── .codex/                   # Codex local environment and bootstrap scripts
├── .pi/                      # Committed Pi config, skills, and built-in extensions
├── agent-workspace/          # Durable agent memory, plans, skills, artifacts, and installs
├── apps/web/                 # TanStack Start application
│   ├── src/routes/           # File-based API and page routes
│   ├── src/lib/pi/           # Pi runtime integration (server.ts, plan-mode.ts, chat-protocol)
│   ├── src/lib/workspace/    # agent-workspace tree and file helpers
│   ├── src/lib/pii/          # PII sanitization module
│   ├── src/lib/logger.ts     # Pino logger with redaction
│   └── src/components/pi/    # Right-panel resources, workspace, and config UI
├── packages/ui/              # Shared React component library
│   └── src/components/
│       └── agent-elements/   # Reusable chat and tool UI
├── docs/                     # Generated and hand-written documentation
├── scripts/                  # Build and utility scripts
└── .github/workflows/        # CI/CD automation

​
Key dependencies

PackagePurpose
@tanstack/react-startFull-stack React framework
@earendil-works/pi-coding-agentPi coding-agent runtime
@earendil-works/pi-aiPi AI primitives
Amazon BedrockPrimary LLM provider
pino + pino-prettyStructured logging
opossumCircuit breaker pattern
zod + @asteasolutions/zod-to-openapiSchema validation & OpenAPI generation
vitest + @playwright/testTesting frameworks
husky + lint-stagedPre-commit hooks

​
Data flow

  1. The browser sends a user message to /api/chat via NDJSON stream.
  2. The server route sanitizes input (PII), logs with correlation IDs, and creates or resumes a Pi session.
  3. The Pi server module invokes Amazon Bedrock through a circuit breaker.
  4. Streaming events (start, delta, tool, done, error) flow back to the client.
  5. The client hydrates messages from the Pi session file on reload and opens supporting resources/workspace panels on demand.
  6. Supporting endpoints expose models, resources, workspace files, sessions, and health checks.
  7. Durable agent context lives in agent-workspace/, including project memory, plans, artifacts, and workspace-installed Pi resources.