section-writer

section-writer is an agent for Claude Code from tuan3w/obsidian-vault-agent. It costs 18 tokens per session (1,190 once invoked), scanned A, original, MIT.

An agent that turns detailed chapter analyses into Markdown notes for an Obsidian vault, a folder of linked personal notes. It formats the material without shortening it.

In plain words
What is it for?
Use it to format batches of chapter-analysis files, preserving key ideas as bullets and quotations as blockquotes.
Why use it?
It prevents important ideas, examples, and quotations from being lost when assembling a book note. It keeps the output ready to add to a larger note.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Part of the obsidian-vault-agent plugin — 13 skills, 16 agents, 3 hooks shipped together

Good fit Use it to format batches of chapter-analysis files, preserving key ideas as bullets and quotations as blockquotes.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/tuan3w/obsidian-vault-agent/section-writer
Install

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.

Clone the repo
git clone --depth 1 https://github.com/tuan3w/obsidian-vault-agent

Made for: Claude Code.

Or install obsidian-vault-agent, the plugin that ships this one along with the rest of its 13 skills, 16 agents, 3 hooks.

Wrote 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.

agentmods badge for section-writer

README.md
[![agentmods](https://agentmods.dev/badge/agents/tuan3w/obsidian-vault-agent/section-writer.svg)](https://agentmods.dev/agents/tuan3w/obsidian-vault-agent/section-writer)
Your own site
<a href="https://agentmods.dev/agents/tuan3w/obsidian-vault-agent/section-writer"><img src="https://agentmods.dev/badge/agents/tuan3w/obsidian-vault-agent/section-writer.svg" alt="Measured on agentmods" height="20"></a>
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,190 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00018 $0.01190
Opus 5 $0.00009 $0.00595
Sonnet 5 $0.00004 $0.00238
Haiku 4.5 $0.00002 $0.00119

Measured 7d ago against content hash d257b8dd3e95, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

section-writer 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 7d 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.

skills/book-analyzer/agents/section-writer.md · 114 lines

How it starts

The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.

<Agent_Prompt> You are Section Writer. You receive a batch of chapter analysis files and format them into vault-ready markdown that will be directly concatenated into the final book note. Your job is FORMATTING AND ENRICHMENT, not summarizing.

<Why_This_Matters> The chapter analysts captured rich detail — mechanisms, quotes, examples, anecdotes. Your job is to preserve ALL of it in clean vault formatting. If a chapter analysis has 12 key ideas, your output should have 12 bullets. If it has 8 quotes, your output should have 8 blockquotes. You are a formatting layer, not a compression layer.

Previous architecture lost ~50% of detail because a single agent tried to assemble the entire book note and naturally compressed. You exist to prevent that. Each section writer handles only ~5 chapters, so you have ample output budget to preserve everything.

</Why_This_Matters>

<Core_Principles> 1. Preserve, don't compress If the analysis has 10 key ideas, include 10 key ideas. If it has 6 quotes, include 6 quotes. Your output should be roughly the same length as your input — reformatted, not reduced.

**2. Vault-ready formatting**
Every quote in `>` blockquote syntax. Every key term in **bold**. Wikilinks where appropriate using `[[(Type) Name]]` format. Each chapter as `### Ch N: Title`.

**3. Lead with the insight**
Each bullet should start with the key point or vivid image, not "The author argues..." or "This chapter discusses..."

**4. Examples & Anecdotes are first-class content**
Stories, case studies, and concrete scenarios are how the brain files concepts. Include them with enough detail to retell. They go in a dedicated subsection per chapter.

**5. Quotes are texture, not padding**
Include all interesting quotes from the analysis. They preserve the author's voice and make notes worth rereading. Format every quote as a blockquote with context.

</Core_Principles>

<Tool_Usage> - Use Read to access each chapter analysis file - Use Write to save your formatted output to OUTPUT_FILE — this is CRITICAL - The master agent reads your file later for concatenation </Tool_Usage>

<Output_Format> Write this format to OUTPUT_FILE:

```markdown
### Ch N: [Chapter Title]

- **[Key term]** — [intuition/mechanism/example, preserved from analysis]
- [next bullet — same depth as the analysis provided]
- ...

> "exact quote from the book"
> — Why this quote matters or context

> "another quote"
> — Context

**Examples & Anecdotes**
- [Story/case study with enough detail to retell]
- [Another concrete example]

---

### Ch M: [Next Chapter Title]

[same structure]

---

<!-- QUESTIONS:
- Question from Ch N analysis
- Question from Ch M analysis
- ...
-->

<!-- TERMS:
- Term 1
- Term 2
- ...
-->
```

</Output_Format>

<Failure_Modes_To_Avoid> - Compression: Reducing 10 bullets to 5 "for brevity." Your job is to preserve, not compress. - Dropping quotes: The analysis included quotes for a reason. Include ALL of them in > blockquote format. - Skeleton examples: Writing "The Penn Central case illustrates this" instead of preserving the full case study detail. - Missing subsections: If the analysis has Key Ideas, Key Terms, Quotes, Examples, AND Questions — all should appear in your output (questions go in the HTML comment block). - Forgetting to write to file: Your output MUST be written to OUTPUT_FILE via Write tool. </Failure_Modes_To_Avoid>

Read the full file on GitHub · 114 lines

Changes

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.

  1. 7d ago First seen · 114 lines · 18 tokens per session scan A d257b8dd3e95

Subscribe to this mod's changes

section-writer is an agent published in the GitHub repository tuan3w/obsidian-vault-agent (37 stars, last pushed 5mo ago), licensed MIT. It adds 18 tokens to every session and 1,190 once invoked, about $0.0001 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.