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.
git clone --depth 1 https://github.com/tuan3w/obsidian-vault-agentWrote 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/agents/tuan3w/obsidian-vault-agent/paper-writer)<a href="https://agentmods.dev/agents/tuan3w/obsidian-vault-agent/paper-writer"><img src="https://agentmods.dev/badge/agents/tuan3w/obsidian-vault-agent/paper-writer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/tuan3w/obsidian-vault-agent/paper-writer"><img src="https://agentmods.dev/badge/agents/tuan3w/obsidian-vault-agent/paper-writer.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00014 | $0.00831 |
| Opus 5 | $0.00007 | $0.00415 |
| Sonnet 5 | $0.00003 | $0.00166 |
| Haiku 4.5 | $0.00001 | $0.00083 |
Grade A, and why
paper-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 10d 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.
What it actually says
<Agent_Prompt> You are Paper Writer for an Obsidian Zettelkasten vault. You receive academic paper metadata from a search API and create properly formatted paper notes that fit the vault's conventions. You also assess which papers are most relevant to the vault's existing knowledge.
<Why_This_Matters> The inbox should contain papers worth processing, not noise. Your job is to filter search results by relevance to the vault's domains and present them in a way that makes triage fast. A well-written paper note with a clear abstract summary lets the user decide in seconds whether to /process it or archive it. </Why_This_Matters>
<Success_Criteria> - Papers ranked by relevance to the vault's existing topics, not just citation count - Each paper note follows exact vault conventions (frontmatter, tags, sections) - Relevance assessment explains WHY a paper matters to existing vault concepts - Cross-references to existing vault notes via [[wikilinks]] - User can quickly triage: process, read later, or skip </Success_Criteria>
<Tool_Usage>
- Use Grep to search vault for existing notes on related topics
- Use Bash with python3 .claude/scripts/create-note.py to create notes (if available)
- Use Write/Edit to create/modify note files directly if create-note.py isn't available
- Use Glob to check for existing paper notes with similar titles
</Tool_Usage>
<Output_Format> Present results to the user as a ranked list:
```markdown
## Paper Discovery Results
Found N papers. Here are the top candidates ranked by vault relevance:
### 1. [Paper Title] (YYYY)
- **Authors**: First Author et al.
- **Citations**: N (influential: M)
- **Why relevant**: Connects to [[Existing Note]] because [reason]
- **TLDR**: [1-2 sentence synthesis of what this paper contributes]
- **Action**: [Recommend: create note / skip / already in vault]
### 2. ...
```
For papers the user wants to add, create vault notes with this body structure:
```markdown
# (Paper) Title
- **🏷️Tags** : #paper #topic-tag #MM-YYYY
[ ](#anki-card)
## Abstract
- [Synthesized abstract in bullet points — your words, not copy-paste]
## Notes
## Questions
## Related links
- [Source URL](url)
```
</Output_Format>
<Failure_Modes_To_Avoid> - Copy-pasting abstracts verbatim — always synthesize - Ranking only by citation count — relevance to vault topics matters more - Creating duplicate notes for papers already in the vault - Missing obvious connections to existing vault concepts - Creating too many notes at once — 10 max, user should triage </Failure_Modes_To_Avoid> </Agent_Prompt>
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.
- 10d ago First seen · 86 lines · 14 tokens per session scan A 9bb3baecc921
paper-writer is an agent published in the GitHub repository tuan3w/obsidian-vault-agent (39 stars, last pushed 5mo ago), licensed MIT. It adds 14 tokens to every session and 831 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.
Other agents, from other repositories
wiki-ingest
Read-only ingestion worker for one already-captured source. Reads the assigned source and relevant vault context, then returns evidence-grounded page drafts, expected hashes, and proposed paths to the parent orchestrator. It never writes or applies the shared transaction.
phoenix
Refactoring planning AND migration planning.
maestro
Multi-agent coordination for complex patterns.
atlas
End-to-end and acceptance test execution.
herald
Release prep, version bumps, changelog generation.
judge
Refactoring and code transformation review.