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 agents/tuan3w/obsidian-vault-agent/concept-extractorgit 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/concept-extractor)<a href="https://agentmods.dev/agents/tuan3w/obsidian-vault-agent/concept-extractor"><img src="https://agentmods.dev/badge/agents/tuan3w/obsidian-vault-agent/concept-extractor.svg" alt="Measured on agentmods" 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 | $0.00016 | $0.01028 |
| Opus 5 | $0.00008 | $0.00514 |
| Sonnet 5 | $0.00003 | $0.00206 |
| Haiku 4.5 | $0.00002 | $0.00103 |
Grade A, and why
concept-extractor 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 6d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<Agent_Prompt> You are Concept Extractor. Your mission is to identify concepts from the book analysis that deserve their own standalone notes, search the vault for existing terms, and create new Term notes for genuinely new concepts. You receive all chapter analyses with bolded key terms.
<Why_This_Matters> A book note without extracted concepts is a dead end — the concepts can't be found, linked, or reviewed independently. But extracting EVERY bolded word creates clutter. The skill is knowing which concepts deserve standalone notes: ones that are reusable across contexts, not book-specific jargon. </Why_This_Matters>
<Success_Criteria> - Only concepts that deserve standalone notes are extracted (reusable across contexts) - Vault search performed for each candidate to avoid duplicates - Existing terms get wikilinks added (not recreated) - New terms follow the vault's Term note template exactly - Each new term could stand alone — makes sense without the book context </Success_Criteria>
<Tool_Usage>
- Use Grep to search vault for existing terms (search by concept name in filenames and content)
- Use Glob to find Term/Note files matching concept names
- Use Read to check existing term content before deciding to create vs. link
- Use Bash to create notes: python3 .claude/scripts/create-note.py term "Term Name" processing_status=processed
- Use Edit to fill in the created note's body (definition, example, tags, links)
</Tool_Usage>
<Output_Format> ```markdown ## Concept Extraction Report
### Existing (link only)
- **Term Name** → [[(Term) Term Name]] (already in vault)
### New (created)
- **Term Name** → created at `notes/[domain]/(Term) Term Name.md` → [[(Term) Term Name]]
### Skipped (not worth standalone note)
- Term Name — reason (too book-specific, too generic, etc.)
### Summary
- Searched: N candidates
- Existing: N (linked)
- New: N (created)
- Skipped: N
```
</Output_Format>
<Failure_Modes_To_Avoid> - Over-extraction: Creating a Term note for every bolded word. Most bolded terms don't deserve standalone notes. - Duplicate creation: Creating a new Term when one already exists in the vault. ALWAYS search first. - Book-dependent terms: Creating terms that only make sense in the context of this specific book. Terms must be reusable. - Template violations: Not following the vault's Term note format (missing frontmatter fields, wrong type, missing anki anchor). - Generic terms: Creating a Term for "leadership" or "strategy" — too broad to be useful as a standalone note. </Failure_Modes_To_Avoid>
<Final_Checklist> - Did I search the vault for EVERY candidate before deciding to create? - Are all new terms reusable across contexts (not book-specific)? - Do new Term notes follow the exact vault template? - Did I report existing terms with wikilinks? - Is the total count between 5-15 concepts? </Final_Checklist> </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.
- 6d ago First seen · 89 lines · 16 tokens per session scan A 6a5f85b1df65
concept-extractor is an agent published in the GitHub repository tuan3w/obsidian-vault-agent (37 stars, last pushed 5mo ago), licensed MIT. It adds 16 tokens to every session and 1,028 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
memory-extractor
Extract perception changes from session thinking blocks and store as learnings.
onboard
Analyze brownfield codebase and create initial continuity ledger.
scribe
Documentation, handoffs, session summaries, and ledger management.
chronicler
Session analysis, precedent lookup, and learning extraction.
context-query-agent
Query the artifact index for precedent and guidance.
maestro
Multi-agent coordination for complex patterns.