Skip to main content
Glama
ForgingStack Watchtower

ForgingStack Watchtower

by jdkabattles

create_watch

Registers a liveness watch on an agent, emailing an alert with a hash-chained receipt if check-ins stop beyond the interval and grace period. Returns a one-time check-in key.

Instructions

Register a liveness watch on an agent. The agent must check in at least every interval_seconds (plus grace); if it goes quiet past that, Watchtower emails an alert and issues a hash-chained receipt. Costs 1 credit. Returns the watch plus a check-in key, shown once — save it: the agent checks in with it, and it cannot be shown again.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "agent_id": "<agent_id>",
      "interval_seconds": "<interval_seconds>",
      "grace_seconds": "<grace_seconds>",
      "notify_email": "<notify_email>"
    },
    "name": "create_watch"
  }
}

Input Schema

NameRequiredDescriptionDefault
agent_idYes
notify_emailYes
grace_secondsYes
interval_secondsYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A3.8/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the burden well: it discloses cost (1 credit), the alerting side effect (email via Watchtower), a hash-chained receipt, and the critical irreversibility that the check-in key is shown only once. Missing are auth/permission prerequisites, duplicate-watch behavior, and failure modes for an invalid agent_id.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Four tight sentences, all front-loaded: what it does, the alert condition, cost, and the return/one-time-key warning. No filler, and the most operationally important warning (save the key) lands at the end where it is likely to be read last and remembered.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a mutating tool with no annotations, no output schema, and 0% parameter documentation, this covers nearly everything an agent needs to call it correctly: the timing contract, cost, and the one-shot secret. Only units, auth requirements, and error behavior are absent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description must compensate for four required parameters. It explains the meaning of interval_seconds and grace_seconds through the check-in window rule but never states units, and agent_id and notify_email are left entirely to their names.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a specific verb and resource: "Register a liveness watch on an agent." The registration semantics clearly separate it from get_watch (retrieval) and check_in (the ongoing ping), though no sibling is named explicitly.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage is implied by the mechanics described — you register a watch when you want email alerting on an agent going silent. There is no explicit statement of when not to use it, or when to prefer get_watch or check_in instead.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools