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-probegit 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-probe)<a href="https://agentmods.dev/skills/burin-labs/harn/harn-probe"><img src="https://agentmods.dev/badge/skills/burin-labs/harn/harn-probe/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-probe"><img src="https://agentmods.dev/badge/skills/burin-labs/harn/harn-probe.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 | $0.00021 | $0.00956 |
| Opus 5 | $0.00010 | $0.00478 |
| Sonnet 5 | $0.00004 | $0.00191 |
| Haiku 4.5 | $0.00002 | $0.00096 |
Grade A, and why
harn-probe 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 — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harn probe
Use this skill when a material claim is uncertain, temporally unstable, high-risk, or expensive to get wrong. Do not probe by reflex when authoritative source or an existing deterministic check already settles the question.
Pair it with [[harn-testing]] when the probe should become a regression and [[harn-de-slop]] when investigation reveals duplicated ownership.
Frame the claim
- State the claim precisely.
- Name the decision the claim affects.
- Name a plausible falsifier.
- Estimate the cost of being wrong.
- Check whether the fact is stable or likely to have changed.
- Identify the authoritative owner or interface.
- Stop if the claim is irrelevant to the requested outcome.
Choose evidence by claim type
- Source claim: inspect the authoritative implementation or registry.
- Contract claim: inspect the type, schema, protocol, or generated artifact.
- Deterministic behavior claim: run a focused test through the public interface.
- Integration claim: exercise the real adapter against a controlled dependency.
- Product claim: run the canonical path end to end.
- Liveness claim: observe progress, interruption, recovery, and terminal state.
- Stochastic quality claim: run multiple trials with a calibrated grader.
- Performance claim: measure representative distributions and resource ceilings.
Probe design
- Use the cheapest probe that can distinguish success from failure.
- Keep one variable changing at a time when possible.
- Capture inputs, environment, version, and current source identity.
- Prefer structured output over screenshots or prose.
- Bound time, concurrency, tokens, and monetary spend.
- Avoid production effects unless explicitly authorized.
- Use test adapters for destructive or expensive dependencies.
- Make the probe repeatable enough to become a regression when useful.
Codebase investigation
- Start from the repository's authoritative registry or interface.
- Trace callers only far enough to validate ownership.
- Search for duplicated policy and parallel implementations.
- Distinguish generated projections from hand-maintained copies.
- Inspect blame, issues, or pull requests when intent is not in current source.
- Check released behavior when source and deployed artifact may differ.
- Record contradictions rather than smoothing them into one story.
- Prefer direct evidence to inferred intent.
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 · 130 lines · 21 tokens per session scan A 58da9644bbf5
harn-probe is a skill published in the GitHub repository burin-labs/harn (21 stars, last pushed today), licensed Apache-2.0. It adds 21 tokens to every session and 956 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
archestra-dev-investigate
Use when investigating Archestra bugs or incidents — staging issues, backend 50x errors, Drizzle failed queries, DB connection pressure, deploy regressions, or Kubernetes/runtime symptoms. Orientation only; defers the process to /investigate.
mcp-cli
Use when running, automating, integrating, or troubleshooting the MCP CLI (mcpa | mcp-ts), including local or remote discovery, authentication, schema inspection, tool calls, and gateway lifecycle errors.
translate
A translation tool for converting user-provided text between languages, with Chinese and English as the default pair.
code_explorer
Explores the repository to locate primary source files, coupled UI components, and test files for bug reports or feature requests.
agentcore-investigation
Investigate Bedrock AgentCore runtime sessions via CloudWatch Logs Insights — resolve session/trace IDs, query OTEL spans, filter noise, build timelines. Use when debugging AgentCore agent sessions, tracing tool calls, or analyzing latency.
cli-e2e
Write, modify, or debug Docker-based Composio CLI end-to-end tests under ts/e2e-tests/cli, including binary invocation, fixture isolation, output assertions, and package manifests. Use for CLI E2E test suites only; use cli-command for CLI source implementation.