Agentsmith

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

🔎 Example output

Real, reproducible output from the tool — runs offline:

$ agentsmith-emit --version
agentsmith 0.1.0

$ agentsmith-emit --help
usage: agentsmith [-h] [--version] [--format {table,json}]
                  {validate,plan,run,init} ...

Config-first multi-agent workflow orchestration.

positional arguments:
  {validate,plan,run,init}
    validate            validate a crew config
    plan                show execution plan (parallel waves)
    run                 validate, plan and execute the workflow
    init                print a runnable starter crew config

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 agentsmith output — reproduce them from a clone.

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

{
"Findings": [
    {
        "id": "123456",
        "title": "Suspicious Network Traffic",
        "description": "Network traffic from unknown IP address",
        "created_at": "2023-02-15T14:30:00Z",
        "updated_at": "2023-02-15T14:30:00Z",
        "labels": ["network", "suspicious"],
        "observables": [
            {
                "type": "ip",
                "value": "192.0.2.1"
            }
        ]
    },
    {
        "id": "789012",
        "title": "Malware Detection",
        "description": "Malware detected on endpoint",
        "created_at": "2023-02-15T14:30:00Z",
        "updated_at": "2023-02-15T14:30:00Z",
        "labels": ["malware", "endpoint"],
        "observables": [
            {
                "type": "file",
                "value": "/path/to/malware"
            }
        ]
    }
]
}

Usage — step by step

agentsmith is config-first multi-agent workflow orchestration: validate a crew config, show its parallel execution plan, and run it.

1. Install (Python 3.10+):

   pip install -e .            # or: pipx install agentsmith
   

2. Scaffold a starter crew config, then save it:

   agentsmith init research-crew > crew.json
   

3. Validate the config and inspect the execution plan (topological parallel waves):

   agentsmith validate crew.json
   agentsmith plan crew.json
   

4. Run the workflow and read the output (per-step outputs + final result); add --format json for machine-readable output:

   agentsmith run crew.json --format json | jq '.steps'
   

5. Gate CI — validate exits 1 on a structurally invalid config, 0 when valid:

   - run: pip install -e . && agentsmith validate crew.json   # non-zero fails the job
   

Contents

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

Why agentsmith?

agent era

agentsmith 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

• ✅ Parse Config
• ✅ Load Config
• ✅ Validate Crew
• ✅ Plan Crew
• ✅ Run Crew
• ✅ Scaffold Config
• ✅ Runs on Linux/macOS/Windows · Docker · devcontainer
• ✅ Ports in Python, JavaScript, Go, and Rust (ports/)

Quick start

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

Example

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

  2 findings · risk score 5 · 38ms

Architecture

flowchart LR
  IN[agent / A2A traffic] --> P[agentsmith<br/>map + analyze]
  P --> OUT[graph + flags]

Use it from any AI stack

agentsmith is interoperable with every popular way of using AI:

• MCP server — agentsmith mcp (Claude Desktop, Cursor, Cognis.Studio, uncensored-fleet)
• OpenAI-compatible / JSON — pipe agentsmith 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

	Cognis agentsmith	CrewAI
Self-hostable, no account	✅	varies
Single command, zero config	✅	⚠️
JSON + SARIF for CI	✅	varies
MCP-native (AI agents)	✅	❌
Polyglot ports (JS/Go/Rust)	✅	❌
Open license	✅ COCL	varies

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

Linux	macOS	Windows	Docker	Cloud
scripts/setup-linux.sh	scripts/setup-macos.sh	scripts/setup-windows.ps1	docker run ghcr.io/cognis-digital/agentsmith	DEPLOY.md (AWS/Azure/GCP/k8s)

Related Cognis tools

• 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
• ragkit — Batteries-included local RAG pipeline — ingest, index, serve
• 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 agentsmith saved you time, star it — it genuinely helps others find it.

Interoperability

{} composes with the 300+ tool Cognis suite — JSON in/out and a shared OpenAI-compatible /v1 backbone. See INTEROP.md for the suite 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.

---