agentmods.dev

runtime-copilot

A set of instructions for operating and diagnosing the Mini Data Engine MCP runtime, a service for running and checking data-engineering workflows.

In plain words
What is it for?
Use it to discover available runtime tools, load default settings, run health checks or regression tests, explain runs, and compare them with baselines.
Why use it?
It provides an organised way to inspect the runtime, check its health, compare results, and investigate failures.

Skill for Claude CodeCodex

0 3mo ago A scan 69 tokens original MIT
View the source file kroq86/data-engineering-runtime-lab
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/kroq86/data-engineering-runtime-lab/runtime-copilot
Any agent
npx skills add kroq86/data-engineering-runtime-lab --skill runtime-copilot
Clone the repo
git clone --depth 1 https://github.com/kroq86/data-engineering-runtime-lab

Made for: Claude Code, Codex.

Per session 69 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 489 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00069 $0.00489
Opus 5 $0.00034 $0.00244
Sonnet 5 $0.00014 $0.00098
Haiku 4.5 $0.00007 $0.00049

Measured yesterday against content hash 63fc15e67d71, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

runtime-copilot scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

codex/skills/runtime-copilot/SKILL.md · 65 lines

What it actually says

Runtime Copilot

Use this skill when working with the Mini Data Engine MCP server from Codex.

The goal is to treat the runtime as a self-describing operational surface, not as a pile of scripts.

Quick start

  1. Call project_tool_catalog to see the current MCP surface.
  2. Call project_get_defaults to load default roots and runtime assumptions.
  3. Use health_check for a fast smoke test.
  4. Use project_run_regression when the user wants a release-style verdict.
  5. Use explain_run and similar_incidents when diagnosing failures.

Working rules

  • Prefer MCP tools before raw shell commands when the capability already exists in MCP.
  • Use shell commands only for setup gaps, file inspection, or changes outside the MCP surface.
  • When summarizing runtime state, mention the exact MCP tools used.
  • If the user asks what is available, answer from project_tool_catalog, not from memory.
  • If the user asks about defaults, answer from project_get_defaults, not from scattered code paths.

For discovery:

  • project_tool_catalog
  • project_get_defaults
  • project_manifest
  • project_capabilities

For health and diagnostics:

  • health_check
  • benchmark_calls
  • scenario_load_test
  • explain_run
  • similar_incidents

For regression and release confidence:

  • project_run_regression
  • project_capture_baseline
  • project_compare_baseline
  • decision_gate

Example requests that should trigger this skill

  • "Show me what this MCP runtime can do."
  • "Check whether the runtime is healthy."
  • "Run the release gate and tell me if this is safe."
  • "Explain why this run failed."
  • "Compare current runtime behavior to the last baseline."

Default answer shape

Keep answers operational and short:

  • what was checked
  • what passed or failed
  • what changed
  • what the next action is
Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 65 lines · 69 tokens per session scan A 63fc15e67d71

Subscribe to this mod's changes

runtime-copilot is a skill published in the GitHub repository kroq86/data-engineering-runtime-lab (0 stars, last pushed 3mo ago), licensed MIT. It adds 69 tokens to every session and 489 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

DatabasesDebuggingETLTestingdata-engineeringdata-pipelinesdatabricksduckdbmcp
Related

Other skills, from other repositories

operator-verify

Bootstrap and verify Operator ETL in one command — Python check, uv install, sync deps, full proof gate. Use on first clone, "does it work", setup, or before any other operator skill.

khaosans/operator-etl · 46 tokens

operator-run

Run Operator ETL locally — MVP demo, pytest, FOIA graph pipeline, Streamlit dashboard. Use when proving the system works or demoing to reviewers.

khaosans/operator-etl · 35 tokens

operator-ship-gcp

Deploy Operator ETL to GCP — Terraform, Docker, Cloud Build, Cloud Run, BigQuery. Use when lifting from local DuckDB MVP to staging/production infrastructure.

khaosans/operator-etl · 41 tokens

okf-maintain

Create, update, and validate the OKF v0.1 knowledge bundle in operator-etl. Use when adding concepts, fixing frontmatter, or preparing a documentation release.

khaosans/operator-etl · 40 tokens

operator-extend

Extend Operator ETL — new sources, domains, gold SQL marts, gov contracts. Use when adding pipelines beyond the FOIA or orders demos.

khaosans/operator-etl · 33 tokens

datacoolie-build

Build, modify, materialize, run, and verify DataCoolie projects. Use for workspace bootstrap, metadata authoring, environment overlays, capability checks, runners, notebooks, custom functions, narrow unsupported adapters, local tests, immutable builds, and project-owned build/CI automation. This is the sole…

datacoolie/datacoolie · 92 tokens