repository-harness is a repository protocol that makes a software codebase easier for coding agents to understand and work on safely. It is for teams using Claude Code, Codex, Cursor, and similar agents who need authoritative documents, durable plans, explicit decision boundaries, and evidence for completed work. Its catalogue entries install the protocol's skills and instructions.
Borrowing it
Nothing to install: this file belongs to hoangnb24/repository-harness. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/hoangnb24/repository-harness/main/.agents/skills/engineering-wisdom/SKILL.mdgit clone --depth 1 https://github.com/hoangnb24/repository-harnessWrote 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/hoangnb24/repository-harness/engineering-wisdom)<a href="https://agentmods.dev/skills/hoangnb24/repository-harness/engineering-wisdom"><img src="https://agentmods.dev/badge/skills/hoangnb24/repository-harness/engineering-wisdom/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/hoangnb24/repository-harness/engineering-wisdom"><img src="https://agentmods.dev/badge/skills/hoangnb24/repository-harness/engineering-wisdom.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00078 | $0.00749 |
| Opus 5 | $0.00039 | $0.00375 |
| Sonnet 5 | $0.00016 | $0.00150 |
| Haiku 4.5 | $0.00008 | $0.00075 |
Grade A, and why
engineering-wisdom 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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Engineering Wisdom
Give practical advice without treating a school of thought as universal law. Repository intent, code, tests, and observed behavior remain authoritative.
Establish Scope
- Read
AGENTS.md,docs/WORKFLOW.md, and only the repository material relevant to the request. - Confirm that the user explicitly invoked this skill. Installation alone is not activation.
- Determine whether the request is advice/review or an authorized change. Keep reviews read-only. For changes, follow repository authority and validation rules before editing.
- Read references/heuristics.md. Select only heuristics that match observed evidence. Read references/sources.md when attribution or the intellectual basis matters.
- When the request crosses a host boundary, inspect the composition root, external input, adapter semantics, shipped artifact, and cumulative state that are relevant. Do not assume isolated core tests prove those boundaries.
Review From Evidence
For each material finding, keep these fields separate:
- Observation: A concrete repository fact with a path, symbol, test, command result, or behavior.
- Heuristic: One named rule of thumb from the catalog and why it may apply.
- Trade-off: Counter-pressure, cost, or a condition under which following the heuristic would make the system worse.
- Proposed repository-owned enforcement: Either
None, or a separately labeled proposal naming the repository owner, authority still required, possible mechanism, and removal condition. Never describe advice as an existing rule unless repository evidence already establishes it. - Verification: A focused observation or executable check that could confirm or falsify the recommendation. Match proof to the claim: test pure policy in isolation, startup at the composition root, and delivery through the artifact users actually load.
Rank findings by user impact and change risk. Prefer one concrete example over several abstract claims.
What ships with it
3 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.
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.
- 9d ago First seen · 79 lines · 78 tokens per session scan A ef5552861a03
engineering-wisdom is a skill published in the GitHub repository hoangnb24/repository-harness (1,215 stars, last pushed 26d ago), licensed MIT. It adds 78 tokens to every session and 749 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
safe-change
Trace, implement, test, and independently review one bounded software change. Use for ordinary feature, bug-fix, or refactor work after the repository is bootstrapped; do not use for broad repository cleanup, production operations, or undefined rewrites.
implementation-loop
Disciplined plan-build-verify loop for implementing a feature or change. Use when building something new or modifying existing behavior, especially across multiple files.
debugging-loop
Systematic reproduce-isolate-fix-verify loop for bugs, test failures, and unexpected behavior. Use when something is broken and the cause is not yet known.
reviewing-code
Reviews code changes for bugs, performance issues, security problems, and best practice violations. Use when reviewing PRs, before committing, after making code changes, or when user asks to review, check, or look over code. Catches N+1 queries, missing error handling, React hooks issues, test coverage gaps, and…
addressing-pr-feedback
Fetches, organizes, and addresses PR review comments from GitHub. Use when user asks to review PR comments, fix PR feedback, check what reviewers said, address review comments, or handle bot suggestions on a pull request. Triggers on "review PR", "fix comments", "PR feedback", "what did reviewers say", "address PR…
critique-theater
Five-dimension design quality review — score the artifact against craft, brand, accessibility, and copy, then fix what falls short before handing it over.