Agent-Harness-Kit: Skill for Claude Code

.agents/skills/governed-review/SKILL.md

governed-review is a skill for Claude Code, Codex from Eduardo-Salvador/Agent-Harness-Kit. It costs 22 tokens per session (444 once invoked), scanned A, original, MIT.

A separate, controlled review process for checking completed coding tasks before they are accepted and integrated.

In plain words
What is it for?
Use it for independent acceptance reviews, objective verification, and integrating work that passes review.
Why use it?
It prevents the implementer’s assumptions and conversation history from influencing the review, while checking the requested behavior and supplied evidence.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions subagents; installed under .agents/ (shared by several agents).

This is Eduardo-Salvador/Agent-Harness-Kit's own configuration. It tells Claude Code and Codex how to work on Agent-Harness-Kit itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Agent-Harness-Kit configures →

Reuse

Borrowing it

Nothing to install: this file belongs to Eduardo-Salvador/Agent-Harness-Kit. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/Eduardo-Salvador/Agent-Harness-Kit/main/.agents/skills/governed-review/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/Eduardo-Salvador/Agent-Harness-Kit

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for governed-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/eduardo-salvador/agent-harness-kit/governed-review/github.svg)](https://agentmods.dev/skills/eduardo-salvador/agent-harness-kit/governed-review)
Your own site
<a href="https://agentmods.dev/skills/eduardo-salvador/agent-harness-kit/governed-review"><img src="https://agentmods.dev/badge/skills/eduardo-salvador/agent-harness-kit/governed-review/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.

agentmods 80×15 button for governed-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/eduardo-salvador/agent-harness-kit/governed-review"><img src="https://agentmods.dev/badge/skills/eduardo-salvador/agent-harness-kit/governed-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 22 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 444 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00022 $0.00444
Opus 5 $0.00011 $0.00222
Sonnet 5 $0.00004 $0.00089
Haiku 4.5 $0.00002 $0.00044

Measured 6d ago against content hash 3f4fe9c9bcb9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

governed-review 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 6d 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.

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.

.agents/skills/governed-review/SKILL.md · 15 lines

What it actually says

Governed review

Load review only for assurance: light|full. assurance: none closes on executor verification regardless of lane. Create the packet because an actual separate reviewer will consume it; never create a ceremonial handoff for same-context work.

Follow ../../../harness/playbooks/review-integration.md and the completed task's review_profile/max_review_rounds. Assurance is automatic and non-blocking: never request human approval, reopen the historical completed node, or stop unrelated ready work.

Round 1 must run under a different identity in a fresh context with no implementer conversation history. When the capability manifest proves spawn_subagent, automatically launch a review subagent; otherwise use a new visible task/chat or manually opened clean context. Never review in the implementer's context or pretend isolation exists. Send only the review packet: pinned executable task SPEC, its explicitly referenced approved authorities, changed paths/diff, immutable handoff, verification/TDD evidence, scoped rules, and read/impact sets. Exclude the original prompt, conversational memory, implementation reasoning, and proposed verdict.

Reconstruct expected behavior from the SPEC before inspecting the implementation. Judge each acceptance criterion against code and independently checked evidence; a prompt/SPEC conflict requires a spec-revision finding, never reinterpretation from memory. Round 1 is proportional to risk; for TDD tasks verify that RED preceded production behavior, failed for the intended reason, GREEN used the identical focused command, and proportional regression passed. Reject fabricated/irrelevant RED or an unjustified non-TDD strategy. Round 2 uses another focused fresh context and pins only the SPEC, linked remediation, prior blocker IDs, correction delta, and proportional regressions. Only evidence-backed acceptance, security/privacy/data, contract, required-runtime, ownership, or material-regression violations create remediation. Record optional improvements as follow-ups. A second rejection forces task/acceptance rewrite, decomposition, or a genuine human product/risk decision. Never request round 3 or expand commit/integration/push/deploy/publication authority.

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. 6d ago Changed 3f4fe9c9bcb9
  2. 12d ago First seen · 15 lines · 22 tokens per session scan A 675fa0d9d4e9

Subscribe to this mod's changes

governed-review is a skill published in the GitHub repository Eduardo-Salvador/Agent-Harness-Kit (6 stars, last pushed 3d ago), licensed MIT. It adds 22 tokens to every session and 444 once invoked, about $0.0001 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.