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/kumaran-is/claude-code-onboardingWrote 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/commands/kumaran-is/claude-code-onboarding/rag-design)<a href="https://agentmods.dev/commands/kumaran-is/claude-code-onboarding/rag-design"><img src="https://agentmods.dev/badge/commands/kumaran-is/claude-code-onboarding/rag-design.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 | $0.00018 | $0.01234 |
| Opus 5 | $0.00009 | $0.00617 |
| Sonnet 5 | $0.00004 | $0.00247 |
| Haiku 4.5 | $0.00002 | $0.00123 |
Grade A, and why
rag-design 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 4d 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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/rag-design — Interactive RAG Architecture Design
You are helping the user design a production RAG system for the project: $ARGUMENTS.
Produce an Architecture Decision Record (ADR) in markdown, saved to docs/adr/NNN-rag-architecture.md (auto-numbered).
Phase 1: Discovery (ask interactively)
Ask these questions one at a time. Wait for the user's answer before continuing. Do not assume:
-
Corpus characteristics
- What kinds of documents? (PDF / Markdown / HTML / DOCX / mixed)
- Approximate corpus size? (thousands / hundreds of thousands / millions of docs)
- Update frequency? (static / daily / hourly / real-time)
- Are there tables, code, images, or scanned docs?
-
Query patterns
- What kinds of questions will users ask? Give 3–5 representative examples.
- Are there proper nouns, IDs, error codes, or rare terms? (BM25 needed)
- Are queries conversational or keyword-style?
- Single-question or multi-step (comparisons, multi-hop)?
-
Users and isolation
- Single-tenant or multi-tenant?
- Access control requirements? (roles, groups, document-level ACLs)
- Regulatory regime? (HIPAA, GDPR, SOC2, none)
-
Latency and scale
- Target p95 latency?
- Expected QPS?
- Cost ceiling per query (if known)?
-
Non-RAG needs
- Are some questions actually about structured data (orders, inventory, account state) that should hit SQL / API instead?
- This is the §19.5 question — flag it explicitly.
Phase 2: Apply the playbook decision framework
Based on answers, walk these decisions and explain each:
- Is RAG the right tool for each query type? Some may route to SQL/API/tool use (§19.5).
- Chunking strategy (chunking decision tree, §10.2)
- Embedding model selection (§14, §15)
- Retrieval method — likely hybrid (BM25 + dense + RRF), justify (§23)
- Filtering / multi-tenancy pattern (§17, §20.2 — always pre-filter)
- Reranking — cross-encoder default (§28)
- Abstention threshold strategy (§30)
- Generation prompt + answer contract (§32)
- Eval approach for v1 — golden set size, metrics (§33–37)
- Operational concerns — observability, cost, ingestion lifecycle
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.
- 4d ago First seen · 140 lines · 18 tokens per session scan A 56efdc241255
rag-design is a command published in the GitHub repository kumaran-is/claude-code-onboarding (35 stars, last pushed 2mo ago), licensed MIT. It adds 18 tokens to every session and 1,234 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 2026-09-03.
Other commands, from other repositories
querying
Query documents from a search index using type-safe filters with support for pagination, sorting, field selection, scoring, and highlighting. Count matching documents efficiently without returning results.
agent
Add an AI agent / RAG backend (@convex-dev/agent) to the Convex app.
ingest
Manually add knowledge to the Weaviate store.
inference.embed
Embed one or more texts into vectors.
vector.batch_exists
Check existence for multiple vector keys.
vector.batch_upsert
Upsert multiple vectors in one itemwise batch.