Mission control evals observability

Skill MN755/Codex-Mission_Control/apps/mcp-server/src/mission_control_mcp_server/_bundled/skills/mission-control-evals-observability

An Open Source Platform For Building Projects With OpenAI's Codex, Claude CLI, And Other Coding LLMs Using Multi-Agent Swarms And Customizable Workflows. Runs On Windows, MacOS, And Linux.

Install

npx -y skills add MN755/Codex-Mission_Control --skill mission-control-evals-observability

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.

What its author says it does

Copied from the file, not written here

Design or run evaluation, tracing, callback, telemetry, and regression-observability workflows for Mission Control agents.

SKILL.md

1.8 KB, as published. Nobody here has run it

Mission Control Evals And Observability

Purpose

Make agent quality measurable with evals, traces, evidence, and regression checks instead of vibes in a trench coat.

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

Use when

The user wants evals or quality gates.
Agent behavior needs traceability.
A workflow needs regression tests or benchmark cases.

Workflow

Ask Mission Control to identify success criteria and failure modes.
Define eval cases, expected outputs, evidence checks, and scoring.
Capture trace points for model calls, tools, approvals, and handoffs.
Summarize results and recommend gates.

Mission Control calls

Tools:

mission_control_start_task
mission_control_get_event_digest

Resources:

mission-control://projects/{project_id}/validation-summary
mission-control://projects/{project_id}/orchestrations/{orchestration_id}/events
mission-control://projects/{project_id}/handoff

User-facing output

Include eval cases, pass/fail status, trace coverage, regressions, and evidence gaps.

Approval behavior

Ask before running costly model/API eval suites or uploading traces externally.

Never do

Do not call an eval meaningful without representative cases.
Do not log secrets in traces.
Do not reduce quality to a single opaque score.

Failure and fallback

If automated evals are unavailable, produce a manual eval rubric and seed cases.

Example invocation

Use Mission Control to design evals for the Ollama worker edit path.