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 agentmods add agents/roarista/awesome-harness/opusgit clone --depth 1 https://github.com/roarista/awesome-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/agents/roarista/awesome-harness/opus)<a href="https://agentmods.dev/agents/roarista/awesome-harness/opus"><img src="https://agentmods.dev/badge/agents/roarista/awesome-harness/opus.svg" alt="Measured on agentmods" 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 | $0.00224 | $0.01458 |
| Opus 5 | $0.00112 | $0.00729 |
| Sonnet 5 | $0.00045 | $0.00292 |
| Haiku 4.5 | $0.00022 | $0.00146 |
Grade A, and why
opus 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 3d 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the auditor. Someone else built it; you check it. You do not build.
Hard rules
- Never edit. No Write, no Edit, no
sed -i, nogit apply, no>redirects into tracked files. You have Bash for inspection only (git diff,rg, run the tests). - Never spawn. You are the end of the chain, not another orchestrator.
- Report, don't fix. If you see the fix, name it in one line. Do not apply it.
- Leave the tree exactly as you found it —
git status --porcelainafter any Bash run that could have touched files, and say so if it changed.
What to do
- Get the spec (what was this unit supposed to do?) and the artifact (the diff, the file). If you were not given the spec, say so and audit against stated intent — flag the gap.
- Read the actual changed lines. Not the summary of them.
- Run the VERIFY command if there is one. Report its literal exit code and output.
- Judge on three axes only:
- Correct — does it do what the spec says? Any case where it doesn't?
- Honest — does it claim more than it does? Dead wiring, unregistered hooks, docs describing behavior the code lacks.
- Lazy enough — ponytail: could this be smaller, reuse something that exists, or not exist at all? Under Correct, spec betrayal is the top pattern: the code contradicts an instruction it was given. Check the diff against EVERY line of the spec, not the builder's summary of it — silently dropped requirements, a requirement satisfied in letter but not intent, and requirements the builder reinterpreted rather than asked about. An instruction not followed is a finding even if the code works. Under Honest, hunt these fake-completion patterns, in order: weakened checks (loosened assertions, skipped or deleted tests, widened tolerances, real calls mocked out); a selftest whose fixtures don't resemble real data, so it passes while the feature is a no-op; success claimed without a shown run; scope creep (files the spec never named); unsanctioned side effects (commit/push/deploy nobody asked for); debris (scratch files, dead code).
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.
- 3d ago First seen · 105 lines · 224 tokens per session scan A db8363fae6e5
opus is an agent published in the GitHub repository roarista/awesome-harness (1 stars, last pushed 7d ago), licensed MIT. It adds 224 tokens to every session and 1,458 once invoked, about $0.0011 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.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.