Agent Security Scanner
Scan AI agents for tool-calling vulnerabilities: prompt leaks, hijacking, injections, and more.
Connector by Mikehzp · GitHub account verified by the MCP Registry
Install
Works with Claude Code, Codex, VS Code, Cursor, Windsurf, Claude Desktop. Read the repository before you give any tool access to your data.
npx agentmarket install connector/mikehzp-agentvuln
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this server can do and what it asks for, from its tool list and install config. Checked Oct 25, 2026.
- Tools
- 4 read · 0 write · 0 delete · 2 not yet classified
- Secrets
- None asked for
- Install scripts
- Not scanned yet
- Runs as
- Python package
Tools by what they do 6
Read from the source code; the running server may list different tools.
Read 4 · Look things up without changing anything.
get_versionlist_attackslist_profileslist_templates
Not yet classified 2 · We could not tell from the name and description.
scan_agentscan_trace
Tools (6)
Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].
get_versionGet the installed agentvuln/agentsec version.
list_attacksList all registered security attack types with metadata. Returns a JSON array of attacks including name, severity, and description for each of the 18+ supported vulnerability types. Returns: JSON string with all available attacks.
list_profilesList available scan profiles and the attacks they include. Profiles are pre-configured attack sets for common use cases: - quick: Fast 5-attack sanity check - daily: Deeper 8-attack routine scan - full: All attacks Returns: JSON string with profile definitions.
list_templatesList available agent simulation templates from the template marketplace. Templates simulate specific agent frameworks (LangChain, Claude Code, etc.) to test how different architectures respond to attacks. Returns: JSON string with template metadata (name, description, version, author).
scan_agentRun a security scan against a live AI agent target. Profiles: quick — 5 critical attacks, ~30s (fast sanity check) daily — 8 attacks, ~2min (routine scan) full — all 18 attacks, ~2min (comprehensive) Targets: deepseek:deepseek-v4-flash, openai:gpt-4o, openrouter:..., hermes-fast, api:https://url/v1:model API keys read from env vars (DEEPSEEK_API_KEY, OPENAI_API_KEY, etc.). Args: target: Target specification (e.g. 'deepseek:deepseek-v4-flash', 'openai:gpt-4o', 'hermes-fast', 'api:https://url/v1:model') profile: Scan profile - 'quick' (5 attacks), 'daily' (8 attack…
scan_traceRun a security scan against an offline agent conversation trace file. Supports LangSmith, LangChain, Claude Code, OpenAI, and generic JSON trace formats. File format is auto-detected. Args: trace_path: Path to the trace file (.json, .jsonl, .ndjson, .claude.json, etc.) attacks: Optional comma-separated list of specific attack names to run. Defaults to all attacks. Returns: JSON string with scan results for the offline trace.
Signals
- Weekly downloads
- 57
- Last release
- Jun 8, 2026 (4 months ago)
- Last package publish
- Jun 8, 2026 (4 months ago)
From the package registry.
Details
- Registry name
io.github.Mikehzp/agentvuln- Version
- 0.4.2
- Auth
- No auth
- Packages
- pypi agentvuln
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- Official MCP Registry
Cite this page
Agent Security Scanner MCP server. Agent Mart. https://agentmart.dev/connectors/mikehzp-agentvuln (last updated 2026-10-24)