RAG Claw
rag-claw
v0.2.0May 22, 2026RAG systems — chunking, embedding, retrieval, and generation with accuracy benchmarks and pipeline visualization
RAG is not "chunk, embed, search, done" — that pipeline fails silently in production. I care about chunk boundary strategies that do not split entities mid-sentence, retrieval precision when your top-10 chunks share 80% overlapping content, reranking that actually moves relevant results up (not shuffles noise), and hallucination rate with and without retrieved context — because if your RAG pipeline is not reducing hallucinations, it is just adding latency. I generate chunking strategy comparisons with entity-preservation scores, retrieval audits with per-query precision/recall, and full-pipeline end-to-end accuracy reports.
OpenClaw
Hermes
Claude Code
Codex
Cursor
Windsurf
Aider
OllamaPRIMARY ACTION
Unlock with ProWhen to Use
- Run local models with private workflows
- Tune inference for local hardware
- Choose effective GGUF variants
- Benchmark practical latency and quality
Compatible Frameworks
OpenClawHermesClaude CodeCodexCursorWindsurfAiderOllama8 TOOLS
Quality Gates
- Complete RAG architecture documented
- Chunking optimized for data type
- Embeddings tuned to target domain
- Generation grounded with citations
- Evaluation framework included
4 GATES DEFINED
Expected Outputs
Native exports per tool
openclaw/AGENTS.mdopenclaw/SOUL.mdopenclaw/TOOLS.md+7 morehermes/skills/flickclaw/rag-claw/SKILL.mdhermes/skills/flickclaw/rag-claw/references/workflow.mdhermes/skills/flickclaw/rag-claw/references/quality-gates.md+2 moreclaude-code/CLAUDE.mdclaude-code/.claude/skills/rag-claw/SKILL.mdclaude-code/.claude/skills/rag-claw/references/workflow.md+3 morecodex/AGENTS.mdcodex/.flickclaw/agents/rag-claw/codex.mdcodex/.flickclaw/agents/rag-claw/workflow.md+2 morecursor/.cursor/rules/flickclaw-rag-claw.mdccursor/.cursor/rules/flickclaw-rag-claw-workflow.mdccursor/.cursor/rules/flickclaw-rag-claw-quality-gates.mdcwindsurf/.windsurf/rules/flickclaw-rag-claw.mdwindsurf/.windsurf/rules/flickclaw-rag-claw-workflow.mdwindsurf/.windsurf/rules/flickclaw-rag-claw-quality-gates.mdaider/CONVENTIONS.mdaider/aider.mdaider/.aider.conf.ymlollama/Modelfileollama/system-prompt.mdollama/template.md+1 moreInstall Commands
Install the FlickClaw CLI, then select your AI agent framework below to get the correct install command.
Step 1: Install CLI (one-time)
npm install -g @flickclaw/cli@latestStep 2: Select Framework
npm exec --yes @flickclaw/cli@latest -- install rag-claw --target openclawDownload as ZIP
Example Prompt
Try this prompt with RAG Claw to see what it can do:
Optimize the local model setup for performance. Benchmark current config and suggest improvements for Chunking Strategy Comparison with Entity Preservation Scores per Method, Retrieval Audit with Per-Query Precision/Recall and Overlap Analysis, Reranking Efficacy Report with NDCG@K Before-After Comparison.Example Output
IllustrativeWhat a typical RAG Claw report looks like:
# RAG Claw — Assessment Report **Project**: ollama-deployment **Context**: a local LLM deployment running Llama 3.1 8B on consumer GPU hardware **Generated**: 2026-07-10 ## Executive Summary The RAG Claw completed its review of ollama-deployment (a local LLM deployment running Llama 3.1 8B on consumer GPU hardware). 3 findings were identified with concrete remediation steps. All quality gates were verified before delivery. ## Findings | # | Severity | Area | Finding | Recommended Action | |---|----------|------|---------|-------------------| | 1 | **P1** | Performance | Inference latency spikes to 8s under concurrency | Enable continuous batching and set max_batch=4 | | 2 | **P2** | Memory | Model consumes 18GB VRAM, headroom insufficient | Switch to Q4_K_M quantization, target <12GB | | 3 | **P2** | Setup | No Modelfile with system prompt defined | Create Modelfile with role, constraints, and templates | ## Quality Gates - [✓] entity_integrity_across_chunk_boundaries - [✓] retrieval_precision_with_overlap_deduplication - [✓] reranking_improvement_over_baseline ## Outputs Generated - **Chunking Strategy Comparison with Entity Preservation Scores per Method**: Included in the report above. - **Retrieval Audit with Per-Query Precision/Recall and Overlap Analysis**: Included in the report above. - **Reranking Efficacy Report with NDCG@K Before-After Comparison**: Included in the report above. - **End-to-End Pipeline Accuracy Report (Hallucination Rate with and without RAG)**: Included in the report above. ## Validation - [x] All quality gates passed (3/3) - [x] 3 findings documented with severity and remediation - [x] 4 output sections generated - [x] Evidence collected and referenced --- *This is an illustrative example output from FlickClaw. Results vary based on project context.*