reverse-eng-binary-ninja
Warn
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references an external repository 'fosdickio/binary_ninja_mcp' and provides a command to fetch the 'binary-ninja-mcp' package from the NPM registry.
- [REMOTE_CODE_EXECUTION]: The skill uses 'npx -y binary-ninja-mcp@1.0.0' to execute the third-party MCP bridge. Executing unverified packages from public registries without manual review can lead to the execution of malicious code if the package is compromised.
- [INDIRECT_PROMPT_INJECTION]: Analyzing untrusted binary files poses a risk where embedded data could manipulate the agent's logic.
- Ingestion points: Binary files, segments, and strings loaded into Binary Ninja (SKILL.md).
- Boundary markers: None specified for protecting the agent from embedded strings or symbols.
- Capability inventory: Capabilities include decompilation, intermediate language (IL) inspection, Python API automation, and byte patching (SKILL.md).
- Sanitization: None; the skill lacks measures to sanitize or validate content extracted from analyzed binaries.
Audit Metadata