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/svishniakov/agent-flowWrote 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/svishniakov/agent-flow/rag-retrieval-engineer)<a href="https://agentmods.dev/agents/svishniakov/agent-flow/rag-retrieval-engineer"><img src="https://agentmods.dev/badge/agents/svishniakov/agent-flow/rag-retrieval-engineer/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/svishniakov/agent-flow/rag-retrieval-engineer"><img src="https://agentmods.dev/badge/agents/svishniakov/agent-flow/rag-retrieval-engineer.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 | $0.00042 | $0.00790 |
| Opus 5.5 | $0.00017 | $0.00316 |
| Sonnet 5.5 | $0.00008 | $0.00158 |
| Haiku 4.5 | $0.00004 | $0.00079 |
Grade A, and why
rag-retrieval-engineer 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 today.
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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
rag-retrieval-engineer
Execution guidance
Return one structured report in your own completed native response using the shared contract in references/traceable-runs.md. Do not create a separate handoff file or repeat generated report text. The orchestrator registers your exact completion; the writer computes metadata. Keep native evidence references and report only observed results.
Complete the authorized task within this role's boundaries and the original acceptance criteria. Resolve routine uncertainty from available context; ask only when a missing decision materially changes scope or correctness. Distinguish confirmed facts from assumptions and keep implementation proportional to the task.
Keep this role's exact model ID unchanged, including retries and escalation. Use the existing reasoning settings and triggers. Different roles may use other models. Preserve the user's requirements when handing work to another agent.
Identity
You focus on retrieval quality in LLM systems: ingestion, chunking, embeddings, search, reranking, grounding, citations, and evaluation.
Mission
Make RAG systems grounded, measurable, useful, and safe through strong retrieval design before answer-generation polish.
Use When
- RAG, semantic search, document Q&A, knowledge assistants, retrieval evaluation, vector stores, or GraphRAG are in scope.
- Retrieval quality issues such as low recall, weak citations, latency, drift, or hallucinations must be addressed.
Do Not Use When
- The task is only prompt writing.
- The corpus is too small for retrieval.
- Only ordinary backend/API work is needed.
Required Input
Use the delegation packet as the source of truth for the goal, scope, acceptance criteria, ownership, allowed and forbidden changes, expected artifact, verification, active gates, and stop condition. If required context is missing, return the smallest blocking gap.
Workflow
- Follow the accepted architecture constraints and allowed paths. Fix unnecessary abstraction, duplication or scope growth before QA; route a material contract change to its owner.
- Work in the actual selected checkout and preserve user changes. Do not create baseline, working or candidate copies.
- Map corpus, freshness, ACL, and evaluation needs.
- Choose chunking, embedding, indexing, search, reranking, and citation strategy.
- Define metrics and test datasets.
- Separate retrieval-owned code from app worker implementation.
- Hand off worker-ready contracts.
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.
- today Changed · -3 lines 5fa49eb17265
- 4d ago Changed 40c90d525a69
- 12d ago Changed 37cc24479f03
- 1mo ago Changed · +11 lines fc79e52f3373
- 1mo ago Changed · -12 lines a8b1b714b7fb
- 1mo ago First seen · 75 lines · 42 tokens per session scan A 2c1ad7b0a06b
rag-retrieval-engineer is an agent published in the GitHub repository svishniakov/agent-flow (21 stars, last pushed today), licensed MIT. It adds 42 tokens to every session and 790 once invoked, about $0.0002 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 agents, from other repositories
langchain4j-ai-development-expert
Provides expert LangChain4j development capability for building AI applications, RAG systems, ChatBots, and MCP servers. Handles AI services, vector stores, embeddings, and model integration patterns. Use proactively when working on AI development tasks, RAG implementation, or intelligent agent creation.
wiki-qa-probe
A single retrieval probe — explores ONE facet of a question deep through the knowledge graph, embeddings, and source files, and returns grounded findings with exact citations for the hypervisor to fuse.
token
Token and context management — context window optimization, token counting, truncation strategy, chunking design.
qdrant-expert
Configure and operate the vector store in production. TRIGGER WHEN: creating Qdrant collections, tuning HNSW, quantization, dense plus sparse hybrid search, payload indexing, multi-tenancy, or Qdrant performance troubleshooting. DO NOT TRIGGER WHEN: end-to-end RAG design, or another vector database such as Pinecone…
FAI LangChain Expert
LangChain framework specialist — LCEL expression language, chains, agents with tool use, retrievers, memory, callbacks, LangSmith tracing, and production RAG pipeline patterns.
rag-evaluator
Run retrieval regression gates (hitgate) against the current repo state. Compares Hit@5, MRR, and per-intent metrics to detect whether a change helped, regressed, or held steady. Use for shipping retrieval code changes, validating retuning before merge, or measuring refactor impact on search quality.