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 agentmods add instructions/easyinplay/harnessed/claude-mdgit clone --depth 1 https://github.com/easyinplay/harnessedWrote 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/instructions/easyinplay/harnessed/claude-md)<a href="https://agentmods.dev/instructions/easyinplay/harnessed/claude-md"><img src="https://agentmods.dev/badge/instructions/easyinplay/harnessed/claude-md.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 | $0.02182 | $0.02182 |
| Opus 5 | $0.01091 | $0.01091 |
| Sonnet 5 | $0.00436 | $0.00436 |
| Haiku 4.5 | $0.00218 | $0.00218 |
Grade A, and why
harnessed CLAUDE.md 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 4d 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 — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.
harnessed — Project CLAUDE.md
Project-specific instructions for Claude Code working on
harnessed. Sister docs: PROJECT-SPEC.md · .planning/STATE.md · docs/WORKFLOW.md Global rules (三层栈方法论 / web search routing / ctx7) live in~/.claude/CLAUDE.md.
Project context
harnessed = AI coding harness 生态的「装配主义包管理器 + composition orchestrator」。不 vendor 上游代码,通过 manifest 描述 install/check + composition skill 编排 ECC / Superpowers / GSD / gstack 等异构上游组件成可执行 workflow。
- Status: 🎯 v0.4.0 SHIPPED & ARCHIVED 2026-05-19 (17/20 phases 85%); v0.5.0 v1.0-RC2 minor STARTING; 🎯 v1.0 GA target 2026-05-22~23
- Stack: Node.js 22 + TypeScript + pnpm + vitest + biome
- Commit safety: NEVER push to remote without user explicit approval (CLAUDE.md commit safety)
- Biome preempt: TS/JS commit 前必跑
pnpm exec biome check --write(3 CI-red recurrences project memory) - Phase cadence: 三层栈完整方法论严格遵守 (gstack governance → GSD orchestration → superpowers brainstorming + karpathy 心法 + mattpocock 招式 + harnessed 自有 completion gate → verify)
gstack skill routing
Use gstack skills for all governance, review, debugging, testing, shipping workflows. Browser automation: always use /browse (or $B from gstack), NEVER use mcp__claude-in-chrome__* tools — they are slow and unreliable.
When the user's request matches one of these patterns, invoke the gstack skill via the Skill tool (not an ad-hoc answer). A false positive is cheaper than a false negative.
Strategy & planning
- Product ideas / "is this worth building" / brainstorm →
/office-hours - Strategy / scope / "think bigger" / "what should we build" →
/plan-ceo-review - Architecture / "does this design make sense" →
/plan-eng-review - Design system / brand / "how should this look" →
/design-consultation - Design review of a plan →
/plan-design-review - Developer experience of a plan →
/plan-devex-review - "Review everything" / full review pipeline →
/autoplan - Save progress / "save my work" →
/context-save - Resume / restore / "where was I" →
/context-restore
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.
- 4d ago First seen · 154 lines · 2,182 tokens per session scan A 85a564f1daed
harnessed CLAUDE.md is an instructions file published in the GitHub repository easyinplay/harnessed (2 stars, last pushed 9d ago), licensed Apache-2.0. It adds 2,182 tokens to every session, about $0.0109 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 2026-08-31.
Other instructions, from other repositories
sd0x-harness CLAUDE.md
Claude Code instructions for sd0xdev/sd0x-harness, covering sd0x-dev-flow — harness engineering for claude code, required checks (stop hook reminded), auto-loop, development rules and footguns.
mumei AGENTS.md
Instructions for iroha924/mumei, covering agent reviewer guidance, what mumei is, setup commands, distribution boundary — flag violations and code style.
mumei CLAUDE.md
Instructions for iroha924/mumei, covering mumei development guide, what mumei is, language policy, bash + jq conventions (essentials) and schemas/ conventions (essentials).
harness-evolver CLAUDE.md
Instructions for raphaelchristi/harness-evolver, covering claude.md, what this is, dependencies, running tools locally and use --no-canary to skip preflight check.
octowiz CLAUDE.md
Instructions for raelli/octowiz, covering octowiz, releasing & deploying, commands, agent skills and issue tracker.
octowiz AGENTS.md
Instructions for raelli/octowiz, covering agent skills, issue tracker, triage labels, domain docs and path validation (dual-validator pattern).