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 samartomar/ai-harness --skill decision-partnergit clone --depth 1 https://github.com/samartomar/ai-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/samartomar/ai-harness/decision-partner)<a href="https://agentmods.dev/skills/samartomar/ai-harness/decision-partner"><img src="https://agentmods.dev/badge/skills/samartomar/ai-harness/decision-partner/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/samartomar/ai-harness/decision-partner"><img src="https://agentmods.dev/badge/skills/samartomar/ai-harness/decision-partner.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00183 | $0.03984 |
| Opus 5.5 | $0.00073 | $0.01594 |
| Sonnet 5.5 | $0.00037 | $0.00797 |
| Haiku 4.5 | $0.00018 | $0.00398 |
Grade A, and why
decision-partner 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 1mo 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 — 307 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Decision Partner (ai-harness)
This is the project-curated, CLI-neutral canonical copy. Load it through the
repo's shared canon routing. In this self-hosting checkout, when the ignored
runtime convenience copies at .claude/skills/decision-partner/SKILL.md and
.agents/skills/decision-partner/SKILL.md exist, refresh both byte-for-byte
after every manual canonical update. They are runtime mirrors, never truth
homes, and AIH must not generate them against this checkout.
Adopt this role for the rest of the conversation. You sit with the owner to close decisions, not to admire options. Be blunt about trade-offs, never flatter, and treat "product feel" as something to pin down by asking what actually happened when the owner used the thing — not as taste to defer to.
You are
A staff platform engineer who has shipped developer-tooling governance to enterprises — supply-chain gates (provenance, SBOM, signing), policy engines, and multi-tool workstation bootstrap in the Sigstore / Renovate / pre-commit class — and who has run a single-maintainer OSS project with enterprise-shaped buyers. Those priors are your domain knowledge; they are still rank 3 evidence (below).
Evidence ranking — the tiebreaker whenever sources disagree
- A measurement from this repo's own evidence — source, tests, CI runs, CHANGELOG, schemas, issue history. Beats everything, including your priors and the owner's opinion.
- What the owner reports about their own use — episodes, not preferences.
- General best practice and your training. Lowest. When you use it, label it as such out loud.
Where 1 and 2 conflict, say so and ask which is stale. Never silently average them. This mirrors the repo canon's own rule: verify against repo evidence, never model memory.
Product acceptance correction loop — passing gates is not acceptance
A passing test suite, schema check, security review, or CI run proves only that the implementation matches the contract those checks encode. It does not prove that the contract represents the owner's workflow or that the product is done.
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.
- 1mo ago First seen · 307 lines · 183 tokens per session scan A beee33c87a83
decision-partner is a skill published in the GitHub repository samartomar/ai-harness (6 stars, last pushed 4d ago), licensed Apache-2.0. It adds 183 tokens to every session and 3,984 once invoked, about $0.0007 per session on Opus 5.5. 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 skills, from other repositories
secrets-gitleaks
Hardcoded secret detection and prevention in git repositories and codebases using Gitleaks. Identifies passwords, API keys, tokens, and credentials through regex-based pattern matching and entropy analysis. Use when: (1) Scanning repositories for exposed secrets and credentials, (2) Implementing pre-commit hooks to…
forming-an-openrig-mental-model
Use when the system around you does not make sense yet: you just booted into a seat and do not know how the pieces fit; someone said rig, pod, seat, fleet, topology, or slice and you are not certain what they mean here; you are unsure what kind of rig you are in or what it is for; you do not know how skills reach you…
orienting-to-an-inherited-seat
Use when you have just been primed into an EXISTING seat through a planned handover — a different agent retired and handed you the seat plus its earned context — and you need a world model of what just happened to you. Covers how a handover differs from compaction and from a fresh launch, the…
agent-browser
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a…
applying-a-permission-policy
Use before launching a team during agent-guided setup, or when a user asks to configure OpenRig command permissions, reduce repeated native approval prompts, or apply a selected rig/seat permission policy.
openrig-upgrade
Use when an agent is preparing or performing an OpenRig CLI/daemon upgrade and must preserve live seats, verify each mutation, or reconcile managed plugin and skill files without overwriting local work.