GUARDPOST

Runtime agent firewall — PII redaction, rate limits, policy enforcement

AI Security & Governance — securing LLMs, agents, and the MCP supply chain.

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

🔎 Example output

Real, reproducible output from the tool — runs offline:

$ guardpost-emit --version
guardpost 0.1.0

$ guardpost-emit --help
usage: guardpost [-h] [--version] [--format {table,json}] <command> ...

Runtime agent firewall: PII redaction, policy enforcement, rate limits.

positional arguments:
  <command>
    scan                scan text/file for PII & policy violations

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

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

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

{
"guardpost": {
"platform": "stix",
"findings": [
{
"id": "1234567890",
"name": "Suspicious Network Traffic",
"description": "Potential malicious activity detected on network 192.168.1.100",
"created_by": "John Doe",
"created_at": "2023-02-20T14:30:00Z"
},
{
"id": "2345678901",
"name": "Malware Detection",
"description": "Virus detected on system with IP address 10.0.0.1",
"created_by": "Jane Smith",
"created_at": "2023-02-21T16:45:00Z"
}
]
}
}

Usage — step by step

1. Install (Python 3.9+):

   pip install guardpost

2. Scan a file (or stdin) for PII and policy violations; redaction is on by default:

   guardpost scan request.txt
   cat prompt.txt | guardpost scan -

3. Tighten the policy: use the strict preset, ban terms, set a rate limit, or tag the principal for audit:

   guardpost scan request.txt --strict --rate-limit 60 --principal svc-bot \
       --ban "internal-secret" --ban "api_key"

4. Read the output: the table shows ALLOWED/BLOCKED, the findings (severity/category/kind/excerpt) and the sanitized text. The process exits 2 when traffic is BLOCKED, 0 when ALLOWED — wire that into a pipeline.

5. Emit machine-readable findings for a gateway / CI:

   cat prompt.txt | guardpost scan - --strict --format json

Contents

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

Why guardpost?

Runtime agent firewall — PII redaction, rate limits, policy enforcement — without standing up heavyweight infrastructure.

guardpost 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

• ✅ Scan Pii

• ✅ Redact

• ✅ Scan Policy

• ✅ Fingerprint

• ✅ Guard

• ✅ Runs on Linux/macOS/Windows · Docker · devcontainer

• ✅ Ports in Python, JavaScript, Go, and Rust (ports/)

Quick start

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

Example

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

  2 findings · risk score 5 · 38ms

Architecture

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

Use it from any AI stack

guardpost is interoperable with every popular way of using AI:

• MCP server — guardpost mcp (Claude Desktop, Cursor, Cognis.Studio, uncensored-fleet)

• OpenAI-compatible / JSON — pipe guardpost 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 protectai/llm-guard, 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 (guardpost 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/guardpost.git"    # pip (works today)
pipx install "git+https://github.com/cognis-digital/guardpost.git"   # isolated CLI
uv tool install "git+https://github.com/cognis-digital/guardpost.git" # uv
pip install cognis-guardpost                                          # PyPI (when published)
docker run --rm ghcr.io/cognis-digital/guardpost:latest --help        # Docker
brew install cognis-digital/tap/guardpost                             # Homebrew tap
curl -fsSL https://raw.githubusercontent.com/cognis-digital/guardpost/main/install.sh | sh

Related Cognis tools

• aegis — AI Agent Permission & Access Auditor — surfaces the lethal trifecta of credentials + injection + reach

• promptmirror — Prompt-injection & indirect-injection scanner for any LLM context input

• ledgermind — Local LLM cost & token forensics proxy with anomaly detection

• adversa — LLM red-team harness — OWASP LLM Top 10 + MITRE ATLAS attack packs

• hallumark — LLM hallucination & grounding auditor for RAG systems

• aicard — Auto-generated NIST AI RMF / EU AI Act Annex IV model & system cards

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 guardpost 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 (licensing@cognis.digital). See LICENSE.