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/reader-levelnpx skills add friedbotstudio/baseline --skill reader-levelgit 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 | $0.00057 | $0.01817 |
| Opus 5 | $0.00028 | $0.00908 |
| Sonnet 5 | $0.00011 | $0.00363 |
| Haiku 4.5 | $0.00006 | $0.00182 |
Grade A, and why
reader-level 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
This skill exists because a page can pass every style check, state every fact correctly, and still be hard to read. The failure is not bad writing. It is writing pitched several grades above the reader, built out of long clauses and Latinate abstractions that each cost a beat to unpack.
Plain language is not dumbing down. The research is consistent: higher-literacy readers prefer plain prose, because it gets them the information faster. An expert reading a dense paragraph is not impressed; they are just spending time they did not need to spend.
The target
Grade 9 on the Flesch-Kincaid scale, and all four measures below. Run:
node .claude/skills/reader-level/score.mjs --target 9 <rendered page>
| Measure | Limit | What it catches |
|---|---|---|
| FK grade | 9 | Sentence length and syllable count |
| Hard words | 10% | Vocabulary the reader has to decode |
| Clause load | 0.4 / sentence | Subordination the reader has to hold |
| Jargon load | 1.2 / sentence | Identifiers per sentence |
One measure alone is gameable, which is why there are four. Flesch-Kincaid is only average sentence length and average syllables per word. A page of twelve-word sentences made of abstractions scores as middle-school and still reads hard, because the reader's cost is carrying unfamiliar concepts, not counting syllables. That is exactly how a page can score grade 7 and feel like a journal article.
The seven moves
Ranked by how much they shift the score.
1. One idea per sentence
Split at every which, so that, because, and when. Two plain sentences beat one correct sentence with a hinge in the middle.
Reach for an epic when one feature decomposes into several pieces that can land separately, and when the scouting and research behind them is shared.
Use an epic when one feature splits into pieces that can ship separately. They also need to share the same background work.
2. Concrete verb over abstract noun
Nouns ending -tion, -ment, -ance, -ity, -ness are usually verbs in disguise. Turn them back.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 159 lines · 57 tokens per session scan A c1269ea2a533
reader-level is a skill published in the GitHub repository friedbotstudio/baseline (11 stars, last pushed 6d ago), licensed Apache-2.0. It adds 57 tokens to every session and 1,817 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
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.