1. Conduid
  2. Agents
  3. Agent Security Bench
MCP server · Agents

Agent Security Bench

by · github.com/sinakazemnezhad/agent-security-bench

Open bench for AI coding-agent ML eval + agent security (prompt injection, jailbreak, tool overreach) with machine-readable receipts

Unclaimed agent-securityjailbreakllm-evalmcpprompt-injectionreceipts
InstallView source
34Low

Scored a month ago · breakdown

About Agent Security Bench

Agent Security Bench is an MCP server published by sinakazemnezhad in the Agents category: open bench for AI coding-agent ML eval + agent security (prompt injection, jailbreak, tool overreach) with machine-readable receipts.

Recent commit activity is the strongest signal that reported issues will get attention.

Install

From source
git clone https://github.com/sinakazemnezhad/agent-security-bench

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Agent Security Bench

Powered by Claude · Grounded in docs

I know everything about Agent Security Bench. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

6 checks have not been run yet.

Releases

v0.5.0v0.5.0 — sec.money_v1 · 16 Aug 2026Agent money control suite. Twin for https://banking.noetfield.com/scorecard
v0.4.0v0.4.0 — Agent Security Evaluation Kit · 16 Aug 2026agent-security-bench 0.4.0 Agent Security Evaluation Kit** — score agent code and refuse unsafe tools. Highlights Wheel-bundled tasks/security/schema/examples (cold \`asb selftest\`) Frontier tasks ml.16–ml.23 (MCP args, HTML injection,…
v0.3.0v0.3.0 — production runner · 16 Aug 2026agent-security-bench 0.3.0 Production-oriented upgrade of the open ML agent + security bench. Runner AST checks (`python_parses`, `ast_has_call`, …) Weighted required checks + `weighted_score` JSON Schema receipt validation (`jsonschema`)…
v0.2.0v0.2.0 · 16 Aug 2026agent-security-bench v0.2.0 Open bench for evaluating AI coding agents on ML tasks and agent security, with machine-readable receipts. Included 10 ML tasks (`tasks/ml/`) Core + extended security suites (prompt injection, jailbreak, tool…
Questions

About Agent Security Bench

How do I install Agent Security Bench?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Agent Security Bench safe to use with an AI agent?

Its trust score is 34 out of 100 (low). Conduid hasn't run static security checks on this repository yet, so review the source yourself before granting it credentials. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Agent Security Bench still maintained?

Yes — the latest release is v0.5.0 (16 Aug 2026), and the last commit was 2 months ago. The repository has 0 stars.