# gemma_worker_run_task

A execute tool on the Mapi Agent Memory MCP server.

Agent View of the PolicyLayer registry record for `gemma_worker_run_task`. HTML page: https://policylayer.com/tools/mapi-agent-memory/gemma-worker-run-task

## Facts

- Tool: `gemma_worker_run_task`
- Server: Mapi Agent Memory (`cabo0m/mapi-agent-memory`) — https://policylayer.com/tools/mapi-agent-memory.md
- Homepage: https://github.com/cabo0m/mapi-agent-memory
- 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": "gemma_worker_run_task",
    "arguments": {}
  }
}
```

## Why gemma_worker_run_task is rated High

The tool name contains 'run_task', which is characteristic of Execute category tools that trigger external operations. Although the description is empty (lowering confidence slightly), the naming convention strongly indicates code or task execution. Given the governance context, misuse could allow an agent to execute arbitrary tasks outside intended parameters.

From the tool's own definition: "Tool name 'gemma_worker_run_task' indicates execution of a task via a worker process. The context of an 'auditable memory and governance' system for AI agents suggests this tool triggers external task execution whose effects depend on arguments provided."

## Use case

AI agents invoke gemma_worker_run_task to trigger actions in Mapi Agent Memory. 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 Mapi Agent Memory:

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

## Other tools on Mapi Agent Memory (251)

- `delete_path` — Destructive — https://policylayer.com/tools/mapi-agent-memory/delete-path.md
- `expire_memory_capture_item` — Destructive — https://policylayer.com/tools/mapi-agent-memory/expire-memory-capture-item.md
- `gemma_worker_cancel_job` — Destructive — https://policylayer.com/tools/mapi-agent-memory/gemma-worker-cancel-job.md
- `rollback_memory_pointer_lifecycle_remediation_execution` — Destructive — https://policylayer.com/tools/mapi-agent-memory/rollback-memory-pointer-lifecycle-remediation-execution.md
- `apply_escalation_reactions` — Execute — https://policylayer.com/tools/mapi-agent-memory/apply-escalation-reactions.md
- `apply_memory_lifecycle_remediation` — Execute — https://policylayer.com/tools/mapi-agent-memory/apply-memory-lifecycle-remediation.md
- `apply_memory_pointer_lifecycle_remediation_execution` — Execute — https://policylayer.com/tools/mapi-agent-memory/apply-memory-pointer-lifecycle-remediation-execution.md
- `backfill_semantic_embeddings` — Execute — https://policylayer.com/tools/mapi-agent-memory/backfill-semantic-embeddings.md
- `gemma_lms_load` — Execute — https://policylayer.com/tools/mapi-agent-memory/gemma-lms-load.md
- `gemma_lms_unload` — Execute — https://policylayer.com/tools/mapi-agent-memory/gemma-lms-unload.md
- `gemma_worker_run_job` — Execute — https://policylayer.com/tools/mapi-agent-memory/gemma-worker-run-job.md
- `git_push` — Execute — https://policylayer.com/tools/mapi-agent-memory/git-push.md
- `git_status` — Execute — https://policylayer.com/tools/mapi-agent-memory/git-status.md
- `preview_memory_pointer_lifecycle_remediation_execution_rollback` — Execute — https://policylayer.com/tools/mapi-agent-memory/preview-memory-pointer-lifecycle-remediation-execution-rollback.md
- `preview_undo_run` — Execute — https://policylayer.com/tools/mapi-agent-memory/preview-undo-run.md
- `query_sql` — Execute — https://policylayer.com/tools/mapi-agent-memory/query-sql.md
- `rollback_memory_hygiene_run` — Execute — https://policylayer.com/tools/mapi-agent-memory/rollback-memory-hygiene-run.md
- `rollback_memory_lifecycle_remediation` — Execute — https://policylayer.com/tools/mapi-agent-memory/rollback-memory-lifecycle-remediation.md
- `rollback_memory_supersession_run` — Execute — https://policylayer.com/tools/mapi-agent-memory/rollback-memory-supersession-run.md
- `run_conflicts_v1` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-conflicts-v1.md
- `run_consolidation_v1` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-consolidation-v1.md
- `run_escalation_check` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-escalation-check.md
- `run_powershell` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-powershell.md
- `run_pytest` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-pytest.md
- `run_sandman_ai` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-sandman-ai.md
- `run_sandman_gemma_hygiene` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-sandman-gemma-hygiene.md
- `run_sandman_quality_hygiene` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-sandman-quality-hygiene.md
- `run_sandman_v1` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-sandman-v1.md
- `run_workshop_action` — Execute — https://policylayer.com/tools/mapi-agent-memory/run-workshop-action.md
- `sandman_memory_chat` — Execute — https://policylayer.com/tools/mapi-agent-memory/sandman-memory-chat.md
- …and 221 more: https://policylayer.com/tools/mapi-agent-memory.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=mapi-agent-memory · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/mapi-agent-memory