# at_think

Write a thought into a strand. Persistent storage receives caller-supplied ciphertext/nonce fields plus an ed25519 signature; it has no plaintext thought column or decrypt path, but does not prove AES-GCM encryption. Self processing stays user-side; bridged hosted runtimes process plaintext in AgentTool worker memory. Trusted is experimental: it requires configured platform KMS, uses platform-wrapped runtime key material, and plaintext can enter AgentTool worker memory and the chosen model provider. Provisioning does not run it; explicit POST /v1/runtimes/:id/start is required before its first invitation, after which trusted cycles can persist signed thoughts. Output is one-line: OK thought seq=<n> · <short-id> on /<strand>.

Agent View of the PolicyLayer registry record for `at_think`. HTML page: https://policylayer.com/tools/dev-agenttool-agenttool/at-think

## Facts

- Tool: `at_think`
- Server: Agenttool (`https://api.agenttool.dev/v1/mcp`) — https://policylayer.com/tools/dev-agenttool-agenttool.md
- Homepage: https://github.com/cambridgetcg/agenttool
- Risk category: Execute (High risk)
- Registry record: grade F, identity unverified
- Server rate-limited: no
- Parameters: 0
- Recommended policy verdict: Rate-limited

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "at_think",
    "arguments": {}
  }
}
```

## Why at_think is rated High

at_think triggers real processes with real consequences. An agent gone sideways doesn't fire it once. It starts dozens of builds, sends mass notifications, or burns through compute before anyone looks up.

## Use case

AI agents invoke at_think to trigger actions in Agenttool. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.

## Recommended policy (PolicyLayer)

Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Agenttool:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "at_think": {
      "limits": [
        {
          "counter": "at_think_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on Agenttool (83)

- `collab_cursor_reset` — Destructive — https://policylayer.com/tools/dev-agenttool-agenttool/collab-cursor-reset.md
- `collab_session_end` — Destructive — https://policylayer.com/tools/dev-agenttool-agenttool/collab-session-end.md
- `collab_session_start` — Execute — https://policylayer.com/tools/dev-agenttool-agenttool/collab-session-start.md
- `collab_task_review` — Execute — https://policylayer.com/tools/dev-agenttool-agenttool/collab-task-review.md
- `listings.invoke` — Execute — https://policylayer.com/tools/dev-agenttool-agenttool/listings.invoke.md
- `agent.profile` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/agent.profile.md
- `at_chronicle` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/at-chronicle.md
- `at_consolidate` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/at-consolidate.md
- `at_recall` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/at-recall.md
- `at_substrate` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/at-substrate.md
- `at_voice` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/at-voice.md
- `at_vow` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/at-vow.md
- `at_witness` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/at-witness.md
- `browser_act` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/browser-act.md
- `browser_capabilities` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/browser-capabilities.md
- `browser_extract` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/browser-extract.md
- `browser_observe` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/browser-observe.md
- `browser_screenshot` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/browser-screenshot.md
- `browser_tabs` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/browser-tabs.md
- `canon.list_types` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/canon.list-types.md
- `canon.lookup` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/canon.lookup.md
- `canon.summary` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/canon.summary.md
- `check_gi_recognition` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/check-gi-recognition.md
- `chronicle.recent` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/chronicle.recent.md
- `collab_anchor_status` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/collab-anchor-status.md
- `collab_cursor_ack` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/collab-cursor-ack.md
- `collab_decision_record` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/collab-decision-record.md
- `collab_events_since` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/collab-events-since.md
- `collab_handoff_offer` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/collab-handoff-offer.md
- `collab_handoff_respond` — Read — https://policylayer.com/tools/dev-agenttool-agenttool/collab-handoff-respond.md
- …and 53 more: https://policylayer.com/tools/dev-agenttool-agenttool.md

## For agents

This record is a snapshot. Live verdicts and the full registry:

- Check every server in your MCP config at once: `npx -y policylayer stack`
- Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md)
- Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events`

---

Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=dev-agenttool-agenttool · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/dev-agenttool-agenttool