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 skills add tpierrain/kenjaku --skill raggit clone --depth 1 https://github.com/tpierrain/kenjakuWrote 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/skills/tpierrain/kenjaku/rag)<a href="https://agentmods.dev/skills/tpierrain/kenjaku/rag"><img src="https://agentmods.dev/badge/skills/tpierrain/kenjaku/rag/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/skills/tpierrain/kenjaku/rag"><img src="https://agentmods.dev/badge/skills/tpierrain/kenjaku/rag.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00137 | $0.01229 |
| Opus 5.5 | $0.00055 | $0.00492 |
| Sonnet 5.5 | $0.00027 | $0.00246 |
| Haiku 4.5 | $0.00014 | $0.00123 |
Grade A, and why
rag 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 1mo 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
rag — the front door to your index
The engine already knows all of this. This skill exists for one reason: it wears the name people reach for. An owner of a RAG-backed brain types
/ragbefore anything else, and until this skill existed the host answered "Unknown command: /rag. Did you mean /run?" — pointing at an unrelated built-in. Nothing was broken; the door had no sign on it.
Principle
Report the status, never compute it. Every number below comes from a tool the vault-rag
server already exposes. This skill routes and translates — it does not derive, estimate, or infer.
A figure that no tool returned does not get said (the repo's don't pretend rule, turned inward).
Procedure
1. "Where is my index at?" → vault_stats
Call the vault_stats tool of the vault-rag MCP server and relay what it returns. It already
carries everything the question is about:
- Documents / chunks indexed, and the breakdown by note type.
- Watcher liveness — whether the live-update watcher is running, and what it last did.
- Embedder identity — which engine vectorizes the notes (fully-local, Ollama, or an API), plus the daily quota when the provider has one (an API); a local embedder has none, and none is displayed.
- Engine + index-schema versions, and whether the index was built against the running one.
Then say it in the owner's words, briefly. The useful translation, not a gloss of every line:
- Documents = your notes. Chunks = the passages they were cut into; search works on those, so a rising chunk count on a stable document count simply means notes grew.
- The watcher running = a note saved in Obsidian is searchable within seconds, with nothing to run by hand. If it is not running, the index has stopped following the notes, so a search can answer from what they said days ago. Say that plainly, then announce the catch-up and run it (step 3) — a 🟡 (ADR 0043): it spends compute and touches no note, so it is not something the owner has to authorise. Close on the veto, in their language: "Say stop if you'd rather I didn't — otherwise I'll go ahead." No question mark, no "shall I".
- A stale index (embedder changed, or schema moved) = search is gated until a re-index; the engine says so itself and offers the re-index. Relay the offer, do not pre-empt it.
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.
- 1mo ago Changed · +4 lines 174a63434cce
- 1mo ago First seen · 69 lines · 137 tokens per session scan A 8dbf992b7bde
rag is a skill published in the GitHub repository tpierrain/kenjaku (92 stars, last pushed 25d ago), licensed Apache-2.0. It adds 137 tokens to every session and 1,229 once invoked, about $0.0005 per session on Opus 5.5. 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
brain-status
Show health of the /brain/ vector store — collection counts (brainraw, brainwiki), indexed vs unindexed files, watcher daemon status, embedding-model consistency, and privacy-mode state. Use when the user says "/brain-status" or asks "is the brain up to date?".
qdrant
Vector search engine for production RAG systems.
chroma
Embedding database for RAG and semantic search.
pinecone
Managed vector DB for production RAG and search.
embeddings
Vector embeddings with HNSW indexing, sql.js persistence, and hyperbolic support. 75x faster with agentic-flow integration. Use when: semantic search, pattern matching, similarity queries, knowledge retrieval. Skip when: exact text matching, simple lookups, no semantic understanding needed.
similarity-search-patterns
Implement efficient similarity search with vector databases. Use when building semantic search, implementing nearest neighbor queries, or optimizing retrieval performance.