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 skills/gigsmart/haiku-method/humanizenpx skills add gigsmart/haiku-method --skill humanizegit clone --depth 1 https://github.com/gigsmart/haiku-methodWhat 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 | $0.00068 | $0.01862 |
| Opus 5 | $0.00034 | $0.00931 |
| Sonnet 5 | $0.00014 | $0.00372 |
| Haiku 4.5 | $0.00007 | $0.00186 |
Grade A, and why
humanize 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 2d 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Humanize: H·AI·K·U voice + AI-tell removal
This skill runs a layered editing pass on blog prose. It is scoped to website/content/blog/**/*.{md,mdx}. Do not invoke it on UI copy, READMEs, doc pages, or paper revisions — those have a different register.
Order of operations
The passes run in this order. Earlier passes win when they conflict with later ones.
- Load the voice rules. Read
.claude/rules/content-voice.mdand.claude/rules/citations.mdbefore touching prose. These define the register and override anything below. - Voice-first edit. Bring the draft into compliance with
content-voice.md: contractions, no hedging, hook in the first two sentences, prose over bullet lists, every claim grounded per the No-Empty-Authority rule, citations linked or paths named percitations.md. - AI-tell sweep. Apply the pattern list in this file — but only where it doesn't fight the voice rules. The conflicts are listed explicitly below; respect them.
- Self-audit. Run the final "what makes this obviously AI?" pass. Answer briefly. Then revise.
Where voice beats the generic humanizer
The standard AI-tell rules are useful but blunt. The voice is specific, and these overrides apply:
- Em-dashes are allowed — particularly in long-form prose where rhythm matters. Don't blanket-replace them with commas. Ban only the AI-cadence em-dash that introduces a marketing flourish ("a seamless experience — designed for you").
- The rule of three is allowed when earned. The voice actively uses three-beat constructions. Strip them only when they're hollow ("efficient, scalable, and powerful"), not when each beat carries weight.
- First-person plural OR singular, not generic. Either "we" (project perspective) or "I" (Jason's specific anecdote). Pick one per post and stay in it. The generic humanizer's "use I when it fits" loses to the explicit register pick.
- Cap "X isn't Y. It's Z." at two uses per article, per
content-voice.md. The humanizer's "negative parallelism" rule aligns here, but the cap is stricter. - Coined phrases stay. When the draft names a pattern ("the continuity contract", "the workshop has two editors"), don't flatten it back to plain language. Coining is a feature, not a tell.
- Contractions are required, not optional.
- No hypothetical numbers, even framed as such. Replace with a real, named, verifiable thing — a commit SHA, a file path, a test case, a CI run — or cut the claim.
- No emojis. Straight quotes. Sentence-case headings.
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.
- 2d ago First seen · 106 lines · 68 tokens per session scan A 3b65511dcbad
humanize is a skill published in the GitHub repository gigsmart/haiku-method (24 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 68 tokens to every session and 1,862 once invoked, about $0.0003 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
mass-line
触发：当你需要收集多方意见、把零散反馈整合成可执行方案，或把方案带回真实使用者/执行者验证时调用；常见信号包括 stakeholder input、user feedback、意见汇总、对齐与验证。 English: Trigger when input must be gathered from many people, synthesized into a clearer plan, and returned to the affected users or executors for validation. Use this skill for a collect-synthesize-validate loop.
workflows
触发：当你面临的任务明显需要多个思想武器协作时调用；常见信号包括：从零启动新项目、攻坚复杂疑难问题、对已有方案进行迭代优化。此 skill 提供标准化的跨 skill 工作流组合，解决"应该先用哪个 skill、怎么衔接"的问题。 English: Trigger when a task clearly requires multiple skills in sequence. Use this skill to select a standard workflow that chains skills together, defines data handoff between steps, and specifies…
concentrate-forces
触发：当多个任务同时争夺时间、注意力、算力或预算，必须确定主攻方向并停止分散用力时调用；常见信号包括优先级过多、资源紧张、推进分散、需要决定先做什么。 English: Trigger when limited resources are being split across too many tasks and one main target must be chosen. Use this skill to concentrate effort, sequence work decisively, and finish a meaningful breakthrough before expanding.
protracted-strategy
触发：当目标长期、任务复杂、资源暂时处于劣势，或短期无法速胜但又不能放弃时调用；常见信号包括 long-term effort、phased plan、endurance、战略耐心、需要分阶段推进。 English: Trigger when the work is long-horizon, difficult, and unlikely to be won quickly. Use this skill to divide the effort into stages, keep strategic confidence, and accumulate small wins into overall victory.
practice-cognition
触发：当你提出了方案、假设或判断，需要通过实践验证、试错迭代或复盘升级认知时调用；常见信号包括 experiment、prototype、validate、iterate、feedback loop。 English: Trigger when an idea, hypothesis, or plan must be tested in practice and improved through iteration. Use this skill to move from action to understanding and back to action in a spiral learning loop.
contradiction-analysis
触发：当问题复杂、存在多个冲突因素、优先级不清，或你不知道应该先解决什么时调用；常见信号包括 trade-off、瓶颈、根因不明、主次不清、多个问题互相牵制。 English: Trigger when a problem contains competing forces, unclear priorities, or no obvious entry point. Use this skill to identify contradictions, isolate the principal contradiction, classify its nature, and choose the right response.