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 daochild/agents-config --skill harness-engineeringgit clone --depth 1 https://github.com/daochild/agents-configWrote 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/daochild/agents-config/harness-engineering)<a href="https://agentmods.dev/skills/daochild/agents-config/harness-engineering"><img src="https://agentmods.dev/badge/skills/daochild/agents-config/harness-engineering/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/daochild/agents-config/harness-engineering"><img src="https://agentmods.dev/badge/skills/daochild/agents-config/harness-engineering.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 | $0.00055 | $0.00698 |
| Opus 5.5 | $0.00022 | $0.00279 |
| Sonnet 5.5 | $0.00011 | $0.00140 |
| Haiku 4.5 | $0.00006 | $0.00070 |
Grade A, and why
harness-engineering 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 today.
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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harness Engineering
The model proposes what to do; the surrounding system controls execution and establishes whether the requested outcome actually happened. Treat reliability as an end-to-end property of context, policy, runtime, verification, and feedback—not as a prompt-only property.
Components
1. Context builder
- Supply the task goal, current state, relevant instructions, constraints, and success criteria.
- Keep always-loaded context concise; retrieve detailed documents and data when relevant.
- Distinguish trusted instructions from user input, retrieved documents, and tool output. Treat the latter as data, not as authority to change policy.
- Do not pass secrets or unrelated personal data into context.
2. Model
- Treat model output as a proposal, not authorization, ground truth, or proof of completion.
- The host controls the agent loop, available tools, and whether a proposal can execute.
- Stop and surface ambiguity, contradictory evidence, or repeated unproductive attempts.
3. Policy gate
- Enforce identity, resource scope, allowed action, and relevant arguments at the tool/action boundary using deterministic host policy.
- Fail closed when authorization cannot be established. Do not rely on model instructions or a model-generated policy judgment to grant access.
- Pause for explicit human approval before actions designated sensitive or consequential by the host's policy.
4. Tools and runtime
- Give each agent only the tools and permissions required for its task.
- Validate tool arguments and constrain file, network, and service access in the host.
- Bound execution with timeouts and call/retry budgets. Stop on exhausted budgets or repeated failures; do not retry side effects unless the operation is safe to repeat.
- Require confirmation or approval for irreversible actions according to host policy.
5. Verification
- Define observable acceptance criteria before execution.
- Verify the resulting artifact or external state with tests, read-back, or another independent check appropriate to the task.
- Report the evidence and any remaining uncertainty. A model's assertion that it succeeded is not verification.
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.
- today First seen · 77 lines · 55 tokens per session scan A 863db49d4445
harness-engineering is a skill published in the GitHub repository daochild/agents-config (16 stars, last pushed today), licensed MIT. It adds 55 tokens to every session and 698 once invoked, about $0.0002 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 2026-10-28.
Other skills, from other repositories
owasp-reviewer-prompt
A paranoid OWASP-Top-10-aware system prompt for AI code review that traces data flow, treats every input as malicious, maps each finding to an OWASP category, and outputs a structured Summary / Findings / Severity / Score block reviewers can act on.
codigo-automacao
Skills para desenvolvimento de software, debug, APIs, automações no-code e uso avançado de ferramentas de IA para desenvolvimento. Use quando precisar de diagnóstico de bugs, construção de APIs, containerização, testes automatizados, integrações ou engenharia de prompts.
dev-flow
Orchestrate an end-to-end product-development workflow built on atomic skills. Reads .workflow/meta.json in a project directory, works out which phase the user is in (idea → PRD → tasks → design → scaffolded → pages → modules → tests → pre-deploy gates → deployed), and delegates to whichever specialist skill owns the…
data-fetching
Read data in a Next.js 16 App Router app the canonical way — async Server Components first, URL searchParams for filter state, use() + when a Client Component needs server data, Route Handlers + SWR/React Query only as a last resort. Server Actions are for mutations, never reads. Use when the user is about to load…
forms
Build or edit any form in a Next.js 16 App Router app through one shared toolkit in lib/forms/, with dirty + valid Save gating — on TanStack Form + Zod (default) or react-hook-form (auto-detected). Use when the user says "form", "edit panel", "create dialog", "settings page", "save button", "dirty state", or reaches…
eve-agent
Scaffold and manage an eve agent (Vercel's filesystem-first agent framework) — inside a Next.js app, as apps/agent in a monorepo, or alone at the repo root when the product has no UI. Use whenever the user wants to create, set up or initialize an eve agent; add a tool, skill, channel, connection, schedule, subagent or…