RAGKIT

pip install cognis-ragkit
ragkit scan .            # → prioritized findings in seconds

🔎 Example output

Real, reproducible output from the tool — runs offline:

$ ragkit-emit --version
ragkit 0.1.0

$ ragkit-emit --help
usage: ragkit [-h] [--version] [--format {table,json}]
              {index,search,ask,stats} ...

Local RAG pipeline: ingest, index, serve.

positional arguments:
  {index,search,ask,stats}
    index               build an index from files/directories
    search              retrieve top-k chunks for a query
    ask                 extractive cited answer for a query
    stats               show index statistics

options:
  -h, --help            show this help message and exit
  --version             show program's version number and exit
  --format {table,json}
                        output format

Blocks above are real ragkit output — reproduce them from a clone.

Sample result format (illustrative values — run on your own data for real findings):

{
"Findings": [
    {
        "ID": "12345",
        "Title": "Suspicious Activity Detected",
        "Description": "An unknown entity accessed our network.",
        "Severity": "High"
    },
    {
        "ID": "67890",
        "Title": "Malware Infection Found",
        "Description": "A virus was detected on a workstation.",
        "Severity": "Medium"
    }
]
}

Usage — step by step

ragkit is a local, dependency-light RAG pipeline: ingest + TF-IDF index, search, and extractive cited answers. Console script: ragkit.

1. Install:

   pipx install ragkit     # or: pip install ragkit
   

2. Build an index from files or directories of .txt / .md documents (written to .ragkit/index.json by default):

   ragkit index ./docs --chunk-size 80 --overlap 20
   

3. Search for the top-k most relevant chunks:

   ragkit search "how do retries work" --top-k 5
   

4. Ask for an extractive, cited answer and read it as JSON (the --format flag is global, before the subcommand):

   ragkit --format json ask "what is the retention policy" --top-k 3 | jq '.answer, .citations'
   

5. Inspect the index in CI/automation to confirm it built and is fresh:

   ragkit --format json stats | jq '.documents, .chunks'
   

Contents

• Why ragkit? · Features · Quick start · Example · Architecture · AI stack · How it compares · Integrations · Install anywhere · Related · Contributing

Why ragkit?

self-host RAG

ragkit is single-purpose, scriptable, and self-hostable: point it at a target, get prioritized results in the format your workflow already speaks (table · JSON · SARIF), gate CI on it, and let agents drive it over MCP.

Features

• ✅ Tokenize
• ✅ Chunk Text
• ✅ Ingest Paths
• ✅ Build Index
• ✅ Save Index
• ✅ Load Index
• ✅ Answer
• ✅ Runs on Linux/macOS/Windows · Docker · devcontainer
• ✅ Ports in Python, JavaScript, Go, and Rust (ports/)

Quick start

pip install cognis-ragkit
ragkit --version
ragkit scan .                       # scan current project
ragkit scan . --format json         # machine-readable
ragkit scan . --fail-on high        # CI gate (non-zero exit)

Example

$ ragkit scan .
  [HIGH    ] RAG-001  example finding             (./src/app.py)
  [MEDIUM  ] RAG-002  another signal              (./config.yaml)

  2 findings · risk score 5 · 38ms

Architecture

flowchart LR
  IN[sources] --> P[ragkit<br/>curate + validate]
  P --> OUT[query / analysis]

Use it from any AI stack

ragkit is interoperable with every popular way of using AI:

• MCP server — ragkit mcp (Claude Desktop, Cursor, Cognis.Studio, uncensored-fleet)
• OpenAI-compatible / JSON — pipe ragkit scan . --format json into any agent or LLM
• LangChain · CrewAI · AutoGen · LlamaIndex — wrap the CLI/JSON as a tool in one line
• CI / scripts — exit codes + SARIF for non-AI pipelines

How it compares

Built in the spirit of RAGFlow, re-framed the Cognis way. Missing a credit? Open a PR.

Integrations

Pipes into your stack: SARIF for code-scanning, JSON for anything, an MCP server (ragkit mcp) for AI agents, and a webhook forwarder for SIEM/Slack/Jira. See docs/INTEGRATIONS.md.

Install — every way, every platform

pip install "git+https://github.com/cognis-digital/ragkit.git"    # pip (works today)
pipx install "git+https://github.com/cognis-digital/ragkit.git"   # isolated CLI
uv tool install "git+https://github.com/cognis-digital/ragkit.git" # uv
pip install cognis-ragkit                                          # PyPI (when published)
docker run --rm ghcr.io/cognis-digital/ragkit:latest --help        # Docker
brew install cognis-digital/tap/ragkit                             # Homebrew tap
curl -fsSL https://raw.githubusercontent.com/cognis-digital/ragkit/main/install.sh | sh

Related Cognis tools

• agentsmith — Config-first scaffolding and orchestration for multi-agent workflows
• skillhub — Local skill registry and installer for AI agents
• toolguard — Runtime allowlist and policy for agent tool-calls
• evalbench — Offline LLM / agent eval harness with regression gates
• memorybank — Portable long-term memory store for agents, exposed over MCP
• promptpack — Versioned prompt / template registry with A/B and rollbacks

Explore the suite → 🗂️ all 170+ tools · ⭐ awesome-cognis · 🔗 cognis-sources · 🤖 uncensored-fleet · 🧠 engram

Contributing

PRs, new rules, and demo scenarios are welcome under the collaboration-pull model — see CONTRIBUTING.md and SECURITY.md.

⭐ If ragkit saved you time, star it — it genuinely helps others find it.

Interoperability

{} composes with the 300+ tool Cognis suite — JSON in/out and a sharedOpenAI-compatible /v1 backbone. See INTEROP.md for thesuite map, composition patterns, and reference stacks.

License

Source-available under the Cognis Open Collaboration License (COCL) v1.0 — free for personal, internal-evaluation, research, and educational use; commercial / production use requires a license ([email protected]). See LICENSE.