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 burin-labs/harn --skill harn-product-qualitygit clone --depth 1 https://github.com/burin-labs/harnWrote 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/burin-labs/harn/harn-product-quality)<a href="https://agentmods.dev/skills/burin-labs/harn/harn-product-quality"><img src="https://agentmods.dev/badge/skills/burin-labs/harn/harn-product-quality/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/burin-labs/harn/harn-product-quality"><img src="https://agentmods.dev/badge/skills/burin-labs/harn/harn-product-quality.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 | $0.00024 | $0.01120 |
| Opus 5 | $0.00012 | $0.00560 |
| Sonnet 5 | $0.00005 | $0.00224 |
| Haiku 4.5 | $0.00002 | $0.00112 |
Grade A, and why
harn-product-quality 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 11d 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harn product quality
Use this skill when Harn behavior reaches a person through a CLI, TUI, IDE, headless integration, or cloud control plane.
Pair it with [[harn-orchestration]] for execution design, [[harn-testing]] for evidence, and [[harn-de-slop]] when surfaces have accumulated parallel policy.
One product
Treat every surface as a projection of one execution contract.
- Harn owns orchestration, lifecycle, replay, evaluation, lineage, and audit.
- Hosts own native presentation, approval UX, mutations, and undo/redo.
- Harn Cloud owns durable remote execution without redefining Harn semantics.
- Keep lifecycle and completion meanings identical across projections.
- Put product-specific rendering behind adapters at a real seam.
- Do not parse prose to recover state the runtime already knows.
- Do not implement a second scheduler or approval engine in a host.
- Fix disagreement at the owner or projection interface.
Canonical path
Design one default path before adding alternatives.
- A new user should discover it without reading architecture documentation.
- The path should demonstrate the product's central value in about a minute.
- Defaults should be safe and useful, not empty configuration homework.
- Show what will happen before consequential work starts.
- Show what is happening while work runs.
- Show what happened and where its evidence lives.
- Keep setup explicit when it cannot be eliminated.
- Test the path from a production-shaped package, not only source checkout.
Control semantics
Control words are events, not suggestions.
stopprevents future work promptly.waitparks without losing resumable state.stand downends delegated or background work cleanly.pivotchanges the active objective without completing stale work.- Surface acknowledgment and resulting state.
- Preserve an auditable reason and initiator.
- Bound long-running tools so cooperative interruption has a turn boundary.
- Make restart and reconnect behavior explicit.
- Test controls during active work, not only at idle.
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.
- 11d ago First seen · 136 lines · 24 tokens per session scan A d2e10ea2179f
harn-product-quality is a skill published in the GitHub repository burin-labs/harn (22 stars, last pushed today), licensed Apache-2.0. It adds 24 tokens to every session and 1,120 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 2026-08-30.
Other skills, from other repositories
gh-issue
Fetch a GitHub issue, create a branch, implement with TDD, and open a PR.
commit
Auto-fix, lint, test, and commit changes with a conventional commit message.
integration-fixture
Create or validate a .test integration fixture under tests/php/Integration/Fixtures.
phel-patterns
Phel language idioms, core library patterns, and testing conventions. Auto-loads when working on Phel source files in src/phel/ or tests/phel/.
phel-repl
Evaluate Phel expressions to verify behavior. Use when you need to test Phel code interactively.
phel-lang
Building applications WITH Phel (Lisp that compiles to PHP). Triggers on .phel files, phel-config.php, phel CLI commands (init, run, repl, test, build), or requests to build a Phel app. Skip when working on the Phel compiler internals (use compiler-guide or phel-patterns instead).