crypto-rsa-attacks
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute specific command-line tools such as
opensslandrsactftoolfor cryptographic analysis and decryption. These are standard tools used within the context of the skill's stated purpose (penetration testing). - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data in the form of RSA public keys, moduli (n), exponents (e), and ciphertexts (c) from potentially untrusted target environments.
- Ingestion points: Extraction of parameters from PEM files, JWT headers, or target system responses (SKILL.md).
- Boundary markers: No specific delimiters or safety instructions are provided to separate untrusted cryptographic parameters from the agent's instructions.
- Capability inventory: The skill utilizes subprocess calls to
opensslandrsactftool, and performs mathematical operations usingsympyandpycryptodome. - Sanitization: No explicit sanitization or validation of the input parameters is mentioned beyond the standard parsing performed by the referenced tools.
Audit Metadata