Shield KYA

Shield KYA: Audit trail and traceability for AI agents that records tool calls and checks them against policy.

Shield KYA provides audit trail and traceability for AI agents. It records every tool call coding agents make on a local dashboard and checks each call against policy. By default, it denies a short list of destructive commands, secret reads, and exfiltration. Local hooks fail open, while real blocking and human approval run on the MCP policy path against the Shield Agent cloud service. It works with any host that speaks Model Context Protocol or OpenAPI.

Shield KYA at a glance

Best for: Software engineer
Runs on: Web, macOS, Windows, Linux, API, VS Code, CLI, self-hosted, open source
Pricing: Free forever under MIT
Released: November 2026
Category: AI coding tool
Website: github.com/the-pixel-boys/shield-kya

Key features

Record every tool call
Check tool calls against policy
Deny destructive commands by default
Provide local dashboard
Support Model Context Protocol
Support OpenAPI
Record advisory events

Use it for

Audit AI coding agent activity
Block destructive agent commands
Track agent tool calls locally
Enforce human approval for agent actions

Alternatives to Shield KYA

iFixAi — Audits agent behavior independently without building manual eval harnesses.
Robusta — Reduces costs and noise by investigating unique incidents instead of raw alerts.
Destructive Command Guard — It prevents AI agents from accidentally running commands that destroy work.
Numbat — It enables local detection and blocking of AI agent actions on endpoints.
Security Audit Skill — Automated multi-stage security auditing with independent verification.
Melra — Governs AI agent tool calls with local policies, approval gates, and verified receipts.
PentAGI — It autonomously manages penetration testing workflows within a secure Docker sandbox.
Stop That Shit — Enforces task boundaries to stop unnecessary AI agent actions.
Laya Guard — Judges every tool call locally on your laptop in ~0.1s with no per-decision billing.
AgentGuard — Get a security and quality assessment of AI skills in seconds without an account.
TaintWall — Enforces deterministic security policies at the tool boundary for AI agents.
Paveo — Enforces deterministic rules on AI agents to prevent destructive actions.

All Shield KYA alternatives