centian
Active·★ 15·Apache-2.0·Updated 2026-07-03
Centian is a crucial layer sitting between AI agents and the systems they interact with, providing security, observability, and control. It captures every agent action, enforces allowed behaviors, and verifies execution against predefined contracts to mitigate risks in production environments.
#AI Agent Security#Agent Observability#Runtime Enforcement#Process Verification#Agent Governance#API Proxy#Audit Trail#Blast Radius Management
01
Features
01Audit trail & observability: Understand what your agents did — and why.
02Realtime context & action guard: Centian governs what enters and leaves the agent's context bidirectionally at runtime.
03Verified execution: Confirm your agents are doing what you actually approved.
04Blast radius management: Exclude catastrophic scenarios by design.
02
Compatibility
Linux
OS
Verified via docs
macOS
OS
Verified via docs
Windows
OS
Verified via docs
Go
Runtime
Verified via docs
Docker
Container
Verified via docs
03
Quick start
1
$ curl -fsSL https://raw.githubusercontent.com/T4cceptor/centian/main/scripts/install.sh | bash
04
Use cases
↳Deploy agents in production safely by mitigating risks.
↳Monitor running agents to quickly notice if something goes wrong.
↳Audit AI decisions for compliance and accountability.
↳Prevent agents from performing destructive actions on critical systems.
05
Alternatives
mastyf.ai★ 21
Open-source AI security platform providing perimeter defense for LLMs and AI agents through swarm analysis, policy enforcement, adversarial testing, and real-time threat detection.
AgentHound★ 452
Offensive security framework for AI agent infrastructure - recon, credential looting, model exfiltration, poisoning, and attack-path analysis across MCP, A2A, gateways, and AI services. BloodHound for the agentic stack.
phantom-secrets★ 18
Stop AI coding agents from leaking your API keys. Local proxy + MCP that swaps real secrets for phm_ tokens — works with Claude Code, Cursor, Windsurf, and Codex.
shellgate★ 15
Self hosted secure gateway for AI agents. One token. Full control. Complete audit trail.
heddle★ 14
Heddle — The policy-and-trust layer for MCP tool servers. Turn YAML configs into validated, policy-enforced MCP tools.
assay★ 13
Open evidence profile for MCP tool actions. Enforce configured tool-call policies and record decisions and observations in offline-verifiable bundles. Optional Linux kernel controls. No hosted backend required.
Comments
Log in to leave a comment
No comments yet. Be the first!