VT-Agent-Firewall

Purpose: Preventing AI agents from invoking unsafe tools or actions without policy checks, approval, and auditability.

Inferred · not functionally tested

Audience: AI security engineers and developers building agentic systems

Inferred · not functionally tested

Functions: agents, monitoring

Inferred · not functionally tested

Interfaces: API: indicated (inferred, not tested) · MCP: indicated (inferred, not tested) · CLI: indicated (inferred, not tested) · Self-hosting: indicated (inferred, not tested)

Recorded constraints: pricing: open_source · license: Apache-2.0 · platforms: CLI · deployment: cli, self_hosted, api_only

Constraint provenance is unknown; confirm requirements with the publisher.

Record sources: pypi.org · github.com. These links do not verify the individual claims.

VT-Agent-Firewall sits in PulseGate's AI & LLM security category. Inferred · not functionally tested: It focuses on preventing AI agents from invoking unsafe tools or actions without policy checks, approval, and auditability. Inferred · not functionally tested: VT-Agent-Firewall is an open-source project aimed at AI security engineers and developers building agentic systems. Basis unknown · not verified: VT-Agent-Firewall is open source under the Apache-2.0 license. Basis unknown · not verified: It runs on the command line and API, and it can be self-hosted.

Valentin Torassa builds and maintains VT-Agent-Firewall, and it first shipped in 2026. Development happens publicly on GitHub with 6 commits in the last 90 days. Inferred · not functionally tested: Key capabilities include request normalization, default-deny policies, and human approval. Inferred · not functionally tested: Catalogued interfaces include an MCP server.

Summary written by a language model from the project’s public pages.