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.
npx skills add Arenukvern/skill_steward --skill mcp-harness-repo-maintainergit clone --depth 1 https://github.com/Arenukvern/skill_stewardWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/arenukvern/skill_steward/mcp-harness-repo-maintainer)<a href="https://agentmods.dev/skills/arenukvern/skill_steward/mcp-harness-repo-maintainer"><img src="https://agentmods.dev/badge/skills/arenukvern/skill_steward/mcp-harness-repo-maintainer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/arenukvern/skill_steward/mcp-harness-repo-maintainer"><img src="https://agentmods.dev/badge/skills/arenukvern/skill_steward/mcp-harness-repo-maintainer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00088 | $0.04376 |
| Opus 5 | $0.00044 | $0.02188 |
| Sonnet 5 | $0.00018 | $0.00875 |
| Haiku 4.5 | $0.00009 | $0.00438 |
Grade C, and why
mcp-harness-repo-maintainer scanned grade C with 2 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 12d ago.
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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
1. Released adopter or CI path: `curl -fsSL https://raw.githubusercontent.com/Arenukvern/skill_steward/main/install.sh | bash`, then `steward <command>`. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
1. Released adopter or CI path: `curl -fsSL https://raw.githubusercontent.com/Arenukvern/skill_steward/main/install.sh | bash`, then `steward <command>`. This is a copy
100% identical to mcp-harness-repo-maintainer — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 280 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Action Contract & Harness Repo Maintainer
Build and maintain repo-local action contracts and harnesses where agents execute and humans steer. The historical mcp- name remains because many adopters arrive through MCP work, but this skill is not MCP-only. For general app, library, tool, plugin, or meta-repo stewardship baselines, use repo-quality-system-lifecycle first; use this skill only when typed actions, probes, benchmarks, or CLI/MCP parity are in scope.
Core principle (action-contract and harness repos)
MCP and CLI are thin interfaces—APIs for agents and CI. Core contains the real logic, schemas, and registries. Adapters parse wire format (argv, MCP JSON-RPC); they delegate immediately.
Agents / CI → CLI ──┐
├──► Core (logic, contracts, tests)
Agents / chat → MCP ──┘
Full layering: core-and-interfaces.md. Parity: every MCP tool must call the same core entrypoint as its CLI twin.
Progressive Automation (Agent-Driven Workflows): Harnesses should let agents turn repeated friction into reviewed, durable capability. If an agent discovers a complex fix or command sequence, it should capture an unknown case or typed action candidate with owner, risk class, inputs, outputs, effects, provenance, and verification. Permanent steward.yaml changes must go through reviewable diffs and validation; do not teach agents to save raw bash permanently from MCP.
Goal-first adoption: The original user goal remains the acceptance check. Tool repair, install work, wrappers, action candidates, evals, and refactors are detours unless they directly solve that goal or preserve a reusable lesson. After two failed repair/setup attempts, stop tool restoration, use a type-native command or portable fallback when possible, record the friction, return to the task, and do not promote from that same detour.
Product experiment ownership: When the goal is visual quality, shader
behavior, loader correctness, renderer throughput, or performance, the product
repo owns the high-throughput experiment runner and oracle. Steward can validate
or summarize an experiment-campaign-summary/v1 artifact after the product loop
has produced captures/metrics. Do not promote a Steward action, benchmark, or
MCP tool as product acceleration unless the product campaign names what surface
changed or was directly proven.
What ships with it
23 files 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.
- evals/cases/adoption-promotion-trigger.yaml 1013 B
- evals/cases/capability-classification-trigger.yaml 929 B
- evals/cases/cold-start-contract-trigger.yaml 530 B
- evals/cases/cooking-recipe-dormant.yaml 418 B
- evals/cases/generational-skeptic-promotion-trigger.yaml 983 B
- evals/cases/goal-first-detour-stop-trigger.yaml 994 B
- evals/cases/interface-split-compress-trigger.yaml 824 B
- evals/cases/native-gate-promotion-trigger.yaml 1.2 KB
- evals/cases/portable-invocation-trigger.yaml 1.0 KB
- evals/cases/protected-local-state-trigger.yaml 560 B
- evals/cases/sibling-layout-trigger.yaml 402 B
- evals/cases/single-transcript-no-promotion.yaml 934 B
- references/cli-mcp-pattern.md 1.8 KB
- references/core-and-interfaces.md 5.1 KB
- references/evals.md 3.1 KB
- references/harness-principles.md 1.5 KB
- references/maintainer-checklists.md 2.7 KB
- references/mcp-production-practices.md 2.5 KB
- references/preferred-tooling.md 4.7 KB
- references/repo-archetypes.md 2.7 KB
- references/sibling-layout.md 2.8 KB
- references/sources.md 468 B
- references/steward-composition.md 1.7 KB
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.
- 12d ago First seen · 280 lines · 88 tokens per session scan C 84b07d7cfa58
mcp-harness-repo-maintainer is a skill published in the GitHub repository Arenukvern/skill_steward (11 stars, last pushed 28d ago), licensed MIT. It adds 88 tokens to every session and 4,376 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). It is 100% identical to mcp-harness-repo-maintainer, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
iflytek-hyper-tts
A text-to-speech tool that turns written text into an MP3 recording. It can use an authorized voice and adjust speaking speed, volume, and pitch.
iflytek-ocr-invoice
An image-reading tool that extracts structured information from Chinese invoices, receipts, bills, and tickets. OCR means turning text in a photo or scan into computer-readable data.
iflytek-pdf-image-ocr
AI-powered OCR service for images and PDF documents using iFlytek's advanced recognition APIs.
iflytek-voiceclone-tts
A tool for creating a speech model from recorded audio and then generating speech that uses the cloned voice. Voice cloning means training software to imitate a person’s voice.
animated-sketch-diagram
A tool for making animated diagrams and flowcharts in a hand-drawn ink style, with paper-like backgrounds, simple icons, and moving dots along connections. It produces a self-contained HTML file and can also create a looping GIF.
iflytek-text-proofread
A tool that checks Chinese writing through iFlytek’s official document-proofreading service. It looks for spelling, punctuation, wording, factual, terminology, and sensitive-content problems.