clawdbot/docs/agent.md

summary, read_when

summary	read_when
Agent runtime (embedded p-mono), workspace contract, and session bootstrap	Changing agent runtime, workspace bootstrap, or session behavior

Agent Runtime 🤖

CLAWDBOT runs a single embedded agent runtime derived from p-mono (internal name: p).

Workspace (required)

CLAWDBOT uses a single agent workspace directory (agent.workspace) as the agent’s only working directory (cwd) for tools and context.

Recommended: use clawdbot setup to create ~/.clawdbot/clawdbot.json if missing and initialize the workspace files.

Full workspace layout + backup guide: docs/agent-workspace.md

If agent.sandbox is enabled, non-main sessions can override this with per-session workspaces under agent.sandbox.workspaceRoot (see docs/configuration.md).

Bootstrap files (injected)

Inside agent.workspace, CLAWDBOT expects these user-editable files:

• AGENTS.md — operating instructions + “memory”
• SOUL.md — persona, boundaries, tone
• TOOLS.md — user-maintained tool notes (e.g. imsg, sag, conventions)
• BOOTSTRAP.md — one-time first-run ritual (deleted after completion)
• IDENTITY.md — agent name/vibe/emoji
• USER.md — user profile + preferred address

On the first turn of a new session, CLAWDBOT injects the contents of these files directly into the agent context.

If a file is missing, CLAWDBOT injects a single “missing file” marker line (and clawdbot setup will create a safe default template).

BOOTSTRAP.md is only created for a brand new workspace (no other bootstrap files present). If you delete it after completing the ritual, it should not be recreated on later restarts.

To disable bootstrap file creation entirely (for pre-seeded workspaces), set:

{ agent: { skipBootstrap: true } }

Built-in tools (internal)

p’s embedded core tools (read/bash/edit/write and related internals) are defined in code and always available. TOOLS.md does not control which tools exist; it’s guidance for how you want them used.

Skills

Clawdbot loads skills from three locations (workspace wins on name conflict):

• Bundled (shipped with the install)
• Managed/local: ~/.clawdbot/skills
• Workspace: <workspace>/skills

Skills can be gated by config/env (see skills in docs/configuration.md).

p-mono integration

Clawdbot reuses pieces of the p-mono codebase (models/tools), but session management, discovery, and tool wiring are Clawdbot-owned.

• No p-coding agent runtime.
• No ~/.pi/agent or <workspace>/.pi settings are consulted.

Peter @ steipete (only)

Apply these notes only when the user is Peter Steinberger at steipete.

• Gateway runs on the Mac Studio in London.
• Primary work computer: MacBook Pro.
• Peter travels between Vienna and London; there are two networks bridged via Tailscale.
• For debugging, connect to the Mac Studio (London) or MacBook Pro (primary).
• There is also an M1 MacBook Pro on the Vienna tailnet you can use to access the Vienna network.
• Nodes can be accessed via the clawdbot binary (pnpm clawdbot in ~/Projects/clawdbot).
• See also skills/clawdbot* for node/browser/canvas/cron usage.

Sessions

Session transcripts are stored as JSONL at:

• ~/.clawdbot/agents/<agentId>/sessions/<SessionId>.jsonl

The session ID is stable and chosen by CLAWDBOT. Legacy Pi/Tau session folders are not read.

Steering while streaming

When queue mode is steer, inbound messages are injected into the current run. The queue is checked after each tool call; if a queued message is present, remaining tool calls from the current assistant message are skipped (error tool results with "Skipped due to queued user message."), then the queued user message is injected before the next assistant response.

When queue mode is followup or collect, inbound messages are held until the current turn ends, then a new agent turn starts with the queued payloads. See docs/queue.md for mode + debounce/cap behavior.

Block streaming sends completed assistant blocks as soon as they finish; disable via agent.blockStreamingDefault: "off" if you only want the final response. Tune the boundary via agent.blockStreamingBreak (text_end vs message_end; defaults to text_end). Control soft block chunking with agent.blockStreamingChunk (defaults to 800–1200 chars; prefers paragraph breaks, then newlines; sentences last). Verbose tool summaries are emitted at tool start (no debounce); Control UI streams tool output via agent events when available.

Configuration (minimal)

At minimum, set:

• agent.workspace
• whatsapp.allowFrom (strongly recommended)

---

Next: Group Chats 🦞