Skip to main content
Glama
Sign In

Outcome-verified watchdog for scheduled agents

by com.textforagents
OverviewDiscussionsAdmin

Server Details

Outcome-verified watchdog for scheduled agents. $0.002/check.

Ownership verified
Status
Healthy
Uptime
100.0% over the last 24 hours
Last Tested
Transport
Streamable HTTP · MCP 2024-11-05
URL
ChatGPT
–
Claude
–

TDQS

B3.2/5.0

Scored across 8 tools

Disambiguation4/5

Most tools target clearly distinct actions: registering jobs, starting runs, checkpointing, completing runs, sweeping, and querying status. The main overlap is between complete_run and verify_now, since both trigger watchdog artifact verification and metered checks, though one is tied to a completion claim and the other is a spot check.

Naming Consistency3/5

The names mix several conventions: verb_noun (complete_run, register_job, start_run), noun/verb forms (checkpoint), noun phrases (last_verified_success), and noun_noun (watchdog_tick). The set remains readable, but there is no single predictable pattern.

Tool Count5/5

Eight tools is well-scoped for a watchdog server covering job registration, run lifecycle, verification, sweeping, and receipt inspection. Each tool has a clear place in the workflow.

Completeness4/5

The core watchdog lifecycle is covered: register a job, start a run, checkpoint progress, claim completion, spot-verify, sweep for stalls/misses, and inspect verdicts/history. Minor gaps remain around canceling active runs, updating or deleting job registrations, and listing jobs/runs directly.

Available Tools

8 tools
checkpointReport run progressBInspect

Report run progress — the watchdog watches progress checkpoints against the job's own wall-clock budget, never process existence. Args: run_id, note?.

Inputs: run_id (required), note.

complete_runComplete a run and verify its outcomeBInspect

Claim completion; the watchdog verifies the declared artifact itself (the claim is never trusted) and issues a signed, hash-chained verdict receipt. Metered check ($0.002 after the one-time 100-check trial / plan rules). Args: run_id, claimed?.

Inputs: run_id (required), claimed.

last_verified_successLast verified successB
Read-only
Inspect

When was this job's last VERIFIED success, and how old is it? Args: job_id.

Inputs: job_id (required).

list_verdictsList verdict receiptsB
Read-only
Inspect

List this operator's signed verdict receipts + receipt-chain re-verification result.

Inputs: no named parameters.

register_jobRegister a scheduled jobCInspect

Register a scheduled job with an outcome contract (file / URL / record, present + non-empty + fresh, optional hash/marker — or a chain contract whose stages' artifacts are the next stage's required input). Args: job_id, contract, stall_budget_s?, due_in_s?, alert via account webhook.

Inputs: job_id (required), contract (required), due_in_s, stall_budget_s.

start_runStart a job runCInspect

Start a run for a job. Args: job_id.

Inputs: job_id (required).

verify_nowVerify a job outcome nowB
Read-only
Inspect

Spot-verify a job's artifact now (independent fetch/hash/read-back by the watchdog). Metered check ($0.002 after trial / plan rules). Args: job_id.

Inputs: job_id (required).

watchdog_tickRun the watchdog sweepAInspect

Run the watchdog sweep: STALLED for active runs with no progress checkpoint inside the job's wall-clock budget, MISSED once per due window with no verified run. Not metered.

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Related MCP Connectors

    Related MCP Servers

    • datasieve-mcp

      gitwormq
      A
      license
      B
      quality
      D
      maintenance
      Enables AI agents to create on-chain and web monitors, dead-man switches, cron triggers, and multi-agent coordination with pay-per-use x402 payments.
      19
      72 npm
      MIT
    • runvouch

      runvouchofficial

      runvouch
      A
      license
      Not graded
      quality
      B
      maintenance
      The watchdog for unattended AI agents: flags MISSED, FAILED, NO_EVIDENCE, RETRY_STORM, BUDGET, DRIFT and STALLED runs of Claude Code routines, OpenClaw, n8n and cron jobs, and alerts via Telegram, Slack or webhook. MIT and self-hostable.
      MIT

    • Confirme

      francistremblay30-netizen
      A
      license
      Not graded
      quality
      C
      maintenance
      Enables AI agents to obtain independent, Ed25519-signed receipts confirming real-world facts: whether a page is online, a domain is legitimate, an email address is deliverable, a heartbeat was declared, or a scheduled task actually ran. Each response includes a verifiable cryptographic receipt, so agents can prove to their users that an asserted outcome was observed by a neutral third party rather than self-attested.
      MIT
    Try in Browser
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    Glama MCP Gateway

    Add one secure layer between your agents and this server.

    Resources