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 guanyang/open-agent-hub --skill harness-engineeringgit clone --depth 1 https://github.com/guanyang/open-agent-hubWrote 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/guanyang/open-agent-hub/harness-engineering)<a href="https://agentmods.dev/skills/guanyang/open-agent-hub/harness-engineering"><img src="https://agentmods.dev/badge/skills/guanyang/open-agent-hub/harness-engineering.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 5.1 | $0.00047 | $0.02297 |
| Opus 5 | $0.00023 | $0.01149 |
| Sonnet 5 | $0.00009 | $0.00459 |
| Haiku 4.5 | $0.00005 | $0.00230 |
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 5d 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.
This is a copy
100% identical to harness-engineering — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 235 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harness Engineering
Harness engineering designs the control system around an agent: what it may edit, how it receives feedback, where it writes state, how failures recover, and who can approve irreversible actions. The harness is the difference between a helpful agent session and an autonomous loop that can run for days without corrupting its objective.
When to Activate
Activate this skill when:
- Building autonomous research or experimentation loops
- Designing an agent environment with locked metrics and editable code or content
- Creating PR-producing or background agents
- Evaluating whether an agent can safely run without frequent human prompts
- Adding novelty, ablation, pruning, rollback, or durable logging to an agent workflow
- Preventing agents from gaming benchmarks, weakening rubrics, or losing state across compaction
Do not activate this skill for adjacent work owned by other skills:
- General quality gates, regression suites, or outcome metrics without autonomous control surfaces:
evaluation. - Tool schemas, response formats, and recovery errors for harness tools:
tool-design. - Project-level task-model fit, pipeline shape, and cost planning:
project-development. - Remote sandbox, warm-pool, and hosted session infrastructure:
hosted-agents.
Core Concepts
Harness Boundary
Separate the agent from the environment it operates inside. The agent proposes actions; the harness defines allowed surfaces, feedback, persistence, and promotion rules.
Use four surface classes:
| Surface | Examples | Rule |
|---|---|---|
| Locked | Eval metric, rubric, validation script, merge policy | Agent may read and propose changes, but cannot score itself with modified rules |
| Editable | Skill draft, experiment file, prompt, config under test | Agent may mutate during the loop |
| Append-only | Results log, research thread, rejected ideas | Agent may append, not rewrite |
| Human-controlled | Merge, production deploy, credentials, destructive operations | Requires explicit human approval |
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.
- 5d ago First seen · 235 lines · 47 tokens per session scan A dde3d8f46d4d
harness-engineering is a skill published in the GitHub repository guanyang/open-agent-hub (961 stars, last pushed today), licensed MIT. It adds 47 tokens to every session and 2,297 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to harness-engineering, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
ac-criteria-validator
Validate acceptance criteria and feature completion. Use when checking if features pass, validating test results, verifying acceptance criteria, or determining feature completion status.
writing-skills
Use when creating new skills, editing existing skills, or verifying skills work before deployment.
iso-write
Implement a written plan using TDD, without committing. Use when invoked as /iso-write [--no-branch | --branch= | --worktree] or handed an implementation plan to build. By default it resumes or cuts a branch derived from the plan filename only when the current branch is a base branch (dev, develop, test, prod, main…
test-coverage-design
Use when deciding what to test, reviewing whether a feature is adequately covered before calling it done, or when tests pass but you're unsure they exercise the real risks. Derives coverage obligations from code + spec, finds gaps (including paths outside the test runner's scope), and designs a minimal…
bestow-curse
Bestow Curse saddles a target with a lasting disadvantage. The real-world version is constraint injection: deliberately adding friction, limitations, or handicaps to see how a system, process, or team adapts. This is the skill of resilience testing through artificial adversity — bandwidth throttling, feature removal…
confusion
Confusion makes targets act randomly and unpredictably. The real-world version is chaos engineering: injecting controlled randomness, unexpected inputs, and edge cases to discover how systems behave when things go wrong. This is fuzzing, monkey testing, and the art of breaking things on purpose so they do not break by…