rag

A command for searching stored memories, conversation transcripts, and information shared between agents by meaning rather than exact words. RAG means retrieving relevant stored information to help answer a question.

In plain words
What is it for?
Use it to search memories and transcripts, filter results by source or agent, inspect matching snippets, and refresh the search index when results are missing.
Why use it?
It helps find related past information when you do not remember the exact wording or file where it appeared.

Command

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.

agentmods
npx agentmods add commands/jansenanalytics/claudex/rag
Clone the repo
git clone --depth 1 https://github.com/JansenAnalytics/claudex
Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 355 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00014 $0.00355
Opus 5 $0.00007 $0.00178
Sonnet 5 $0.00003 $0.00071
Haiku 4.5 $0.00001 $0.00036

Measured yesterday against content hash 9fa4f2688c85, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 yesterday.

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.

commands/rag.md · 20 lines

What it actually says

Semantic search across Claudex's memory store, session transcripts, and cross-agent context. Requires OPENAI_API_KEY (used for embeddings).

  1. Run node --experimental-sqlite ~/.claude-agent/scripts/memory-search.cjs --search "$ARGUMENTS" --limit 10. If it errors on a missing OPENAI_API_KEY, say so plainly — that's the fix, not a bug.
  2. If zero results, retry once without any --source/--agent filters. Still nothing → suggest node --experimental-sqlite ~/.claude-agent/scripts/memory-search.cjs --index --incremental to refresh the index, then re-run.
  3. Present the top hits, Telegram-friendly, one bullet each:
    • [memory|session|cross-agent] source tag + score (e.g. 0.82)
    • 1-2 line snippet of the matched content
    • file/line ref in code (e.g. memory/2026-05-12.md:14)
  4. End with: "Drill into a hit?" — offer to open any one and read it in full.

Filter options to mention if useful (don't run unprompted): --source memory|session|cross-agent to scope, --agent kite|argus for a specific sister agent's context.

Output: bullets only, no tables, no preamble. Lead with a one-line verdict (✅ N hits / ⚠️ weak matches / ❌ nothing).

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. yesterday First seen · 20 lines · 14 tokens per session scan A 9fa4f2688c85

Subscribe to this mod's changes

rag is a command published in the GitHub repository JansenAnalytics/claudex (5 stars, last pushed 2mo ago), licensed MIT. It adds 14 tokens to every session and 355 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 2026-08-31.