Quick Commands

PolitiClaw registers a handful of "quick commands" that bypass the agent entirely. They return canned, deterministic text — useful when you want a fast answer without paying model-token cost or waiting on inference.

These complement the tools, they do not replace them. Anything that needs interpretation, drafting, or alignment scoring still flows through the agent.

When to use a command instead of a tool

Use a quick command when you want:

• a navigation hint (/politiclaw-help)
• the next onboarding step and copyable agent prompt (/politiclaw-setup)
• a snapshot of saved state without rerunning the doctor tool (/politiclaw-status)
• a structured health check (/politiclaw-doctor)
• a key inventory and what each unlocks (/politiclaw-keys)
• the plugin version and the OpenClaw runtime floor (/politiclaw-version)

Use a tool when you want narrative, scoring, drafting, or anything that benefits from the agent's reasoning — politiclaw_election_brief, politiclaw_score_representative, politiclaw_draft_outreach, etc.

The six commands

/politiclaw-help

Lists the core (Tier 1) tools alongside the other quick commands. Generated from the same tool registry the published reference reads, so the names and one-line summaries stay in sync automatically.

/politiclaw-setup

Shows the next setup step and a copyable prompt for the agent. It does not mutate state or call tools. Before setup, it tells you to ask the agent to call politiclaw_configure with your street address. Mid-flow, it reads the onboarding checkpoint and prints the next continue prompt. After setup, it points to /politiclaw-status, /politiclaw-doctor, and common follow-up tools.

/politiclaw-status

Snapshot of the plugin's saved state: address (state and zip), monitoring mode, accountability setting, action prompting, issue stance count, and how many of the supported API keys are configured. Reads the plugin's private SQLite database directly — no model invocation.

/politiclaw-doctor

Runs the same checks as the politiclaw_doctor tool — schema migrations, SQLite integrity, preferences, accountability mode, API keys, reps cache, cron jobs, and skill overrides — and prints them as a flat list with [ok] / [warn] / [fail] markers and an actionable hint per non-OK check. Use this for a fast, no-cost health check; use the tool for the agent-mediated walkthrough.

/politiclaw-keys

Lists every supported API key with its requirement (required vs optional), current state (set vs not set), and what each unlocks. Use this when you're trying to figure out why a particular feature isn't working — if the key is not set, the feature won't be wired.

/politiclaw-version

Three lines: the PolitiClaw plugin version (from package.json), the plugin API floor (openclaw.compat.pluginApi in package.json), and the minimum OpenClaw host version (openclaw.install.minHostVersion). Use this when filing bug reports or debugging compatibility.

Where the canned content comes from

Each command's text is generated from existing source-of-truth metadata so it can't drift from the rest of the docs:

• /politiclaw-help reads REGISTERED_POLITICLAW_TOOL_DOCS and TOOL_AUDIT_ENTRIES (the same tier table that drives apps/docs/maintainers/tool-surface.md).
• /politiclaw-setup, /politiclaw-status, and /politiclaw-doctor read the live SQLite database via the plugin's storage helpers, plus getPluginConfig() for the API key state where needed.
• /politiclaw-keys and the doctor's API-key check read a single API_KEY_FLAGS constant exported from src/domain/doctor/checks.ts.
• /politiclaw-version reads the plugin's own package.json at runtime.