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/friedbotstudio/baseline/prosenpx skills add friedbotstudio/baseline --skill prosegit clone --depth 1 https://github.com/friedbotstudio/baselineWhat 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 | $0.00106 | $0.02377 |
| Opus 5 | $0.00053 | $0.01189 |
| Sonnet 5 | $0.00021 | $0.00475 |
| Haiku 4.5 | $0.00011 | $0.00238 |
Grade A, and why
prose 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 yesterday.
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 — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are executing a decision the main context has already made: "produce this specific prose, for this audience, in this register, grounded in this source material." You compose. You do not invent facts, pick the register, or expand scope.
The skill is invalid without an explicit Skill(humanizer) tool call
This is the load-bearing rule. The other rules support it.
- Every prose deliverable produced by this skill must end with the model issuing a
Skill(humanizer)tool call against the draft, and using humanizer's output verbatim as the final text. - Reading humanizer's patterns from memory and rewriting "in humanizer's spirit" does not satisfy this rule. The Skill tool call must occur in the same turn as the prose.
- "I know these patterns, I'll skip the call to save tokens" is the exact failure this rule prevents. Skip is forbidden even for a single sentence.
- The receipt line at the end of your output must reference the turn-local humanizer Skill tool call. If you cannot honestly write that line, the deliverable is not done.
Before you draft — load this checklist
Hold these patterns in active context for the whole drafting pass. Drafting against the checklist is cheaper than rewriting after.
Forbidden in human-facing prose:
- Em-dash overuse. Treat the em dash as expensive. Maximum one em dash per paragraph. Two em dashes in the same paragraph is always wrong; stacking em dashes around a parenthetical (
A — X — B) is always wrong. - Sentence-fragment stacking. Three or more short fragments in a row read as AI rhythm. "Skills run here. One worker there. Discipline through composition." — that pattern is the tell. Vary length. Break the rhythm.
- Sloganeering in body copy. "Placement is policy." "The audit is the contract." "Discipline through composition." Headline rhythms in paragraph copy are AI signatures. State the claim plainly and move on.
- Tagline echo. If the headline contains a phrase, the body paragraph should not repeat that phrase. Echoing reinforces the slogan and feels engineered.
- Negative parallelism. "It's not just X, it's Y." "What this is not: …. It is also not …." Cut these structures unless one specific instance is genuinely the cleanest expression.
- Rule of three. Forced triplets that round out a list to three items for cadence. If you have two real items, write two.
- AI vocabulary. crucial, pivotal, leverage, robust, comprehensive, seamless, holistic, foster, navigate, journey, harness (verb), unleash, cutting-edge, game-changing, paradigm, synergy, delve, tapestry, testament, underscore, landscape (abstract), vibrant, in the heart of, nestled.
- Vague attributions. "Industry experts believe", "research suggests", "many would argue".
- Filler hedges. "It is important to note that", "in order to", "at this point in time", "due to the fact that".
- Generic positive endings. "Exciting times ahead." "The future looks bright." "A major step forward."
- Bolded inline-header lists (
- **User Experience:** …) and emoji decoration. Cut both. - Curly quotes (
“”‘’). ASCII straight quotes only.
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.
- yesterday First seen · 129 lines · 106 tokens per session scan A 6ded045c1424
prose is a skill published in the GitHub repository friedbotstudio/baseline (11 stars, last pushed 5d ago), licensed Apache-2.0. It adds 106 tokens to every session and 2,377 once invoked, about $0.0005 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
dev-standards
Enforces development workflows, quality gates, coding standards, and release processes for the deterministic-agent-control-protocol project. Use when implementing features, fixing bugs, refactoring architecture, adding integrations, updating policies, writing tests, updating documentation, or preparing releases.
code-review-with-lsp
Code review with LSP-powered code intelligence. Uses MCP tools (diagnostics, hover, references, definition, symbols) for semantic code understanding, not just text grep.
i18n-check
国际化完整性检查。检查翻译 key 是否缺失、硬编码文本、locale 文件一致性。.
vue-best-practices
Vue 2/3 代码规范检查。包括组件命名、Props 校验、Composition API 规范等。.
python-review
Python 遗留代码审查：bare except、SQL 注入、反序列化、密钥、调试输出.
rust-review
Rust 服务审查：panic、SQL 注入、密钥、错误吞没、遗留标记.