agentsclimarketplace
Sign in

Mission control agent stuck

Skill MN755/Codex-Mission_Control/.codex/skills/mission-control-agent-stuck

Diagnose a stuck Mission Control agent. Use when an agent shows no output, repeats errors, times out, asks for the same approval repeatedly, or stops making progress events.From its SKILL.md

Install
npx -y skills add MN755/Codex-Mission_Control --skill mission-control-agent-stuck

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.0 KB, 378 tokens by cl100k_base, as published. Nobody here has run it

Mission Control Agent Stuck

Purpose

Diagnose a stuck agent and surface recovery options through Mission Control.

The Codex chat agent is not the Mission Control Manager. It is the bridge between the user and the Mission Control Manager.

Use when

  • A specific agent looks frozen.
  • Progress events stopped.
  • Repeated errors or approvals suggest a stuck loop.

Workflow

  1. Read agent status, event digest, diagnostics, and pending decisions.
  2. Identify the stuck signal: no output, repeated error, timeout, repeated approval request, or no progress events.
  3. Ask Mission Control for recovery options when available.
  4. Present retry, reassign, split task, pause, or debug-agent options if backed by Mission Control.

Mission Control calls

Tools:

  • mission_control_get_event_digest
  • mission_control_get_status
  • mission_control_request_recovery_plan

Resources:

  • mission-control://projects/{project_id}/agents
  • mission-control://projects/{project_id}/diagnostics
  • mission-control://projects/{project_id}/pending-decisions

User-facing output

  • Show the stuck signal, likely cause, affected agent, and safe recovery options.
  • Keep the explanation focused on actionability.

Approval behavior

If recovery means retrying commands, reassigning work, or spawning extra debugging capacity, keep those behind Mission Control approvals.

Never do

  • Do not restart or replace the agent from chat by assumption.
  • Do not ignore a repeated approval loop.
  • Do not claim the agent is healthy if the signals are stale.

Failure and fallback

If dedicated stuck-agent logic is not exposed yet, infer the stuck pattern from status, diagnostics, and event digest, then recommend the recovery-plan flow.

Example invocation

This Mission Control agent looks stuck. Diagnose it.

What ships with it: 1 file

141 B alongside SKILL.md

agents/

Gives 0 of the 12 instructions most context ai engineering skills give in 378 tokens

Counted across 1,193 of the 1,976 authors here whose files we hold, read 2026-08-07

  • Dispatch a fresh implementer subagent per taskin 48 of 1193, across 19 files
  • Dispatch a final code reviewer after all tasksin 33 of 1193, across 8 files
  • Provide full task text to the subagentin 30 of 1193, across 9 files
  • Review spec compliance before code qualityin 27 of 1193, across 10 files
  • Make the hook script executablein 26 of 1193, across 8 files
  • Re-snapshot after navigation or DOM changesin 25 of 1193, across 19 files
  • Read files before editing themin 22 of 1193, across 11 files
  • Answer subagent questions before proceedingin 22 of 1193, across 7 files
  • Mark task complete in TodoWrite after approvalin 22 of 1193, across 6 files
  • Merge hook into existing settingsin 21 of 1193, across 3 files
  • Ask if installation is global or projectin 20 of 1193, across 2 files
  • Copy the hook script to target locationin 20 of 1193, across 2 files

Said here and by no other author read

  • Read agent status and event digest
  • Identify the specific stuck signal
  • Request recovery options from Mission Control
  • Present safe recovery options to the user
  • Keep explanations focused on actionability
  • Infer stuck patterns if dedicated logic is missing

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,144. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.