rsa-attack-techniques
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for the agent to execute external cryptographic binaries such as
RsaCtfTool,yafu, andmsieveto perform integer factorization and ciphertext decryption. - [EXTERNAL_DOWNLOADS]: The skill references
factordb.com, which is a well-known community resource for looking up integer factorizations. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface by ingesting and processing user-supplied cryptographic parameters to execute math scripts.
- Ingestion points: User-provided RSA parameters including the modulus (n), public exponent (e), and ciphertext (c) used in the script templates in
SKILL.mdandRSA_ATTACK_CATALOG.md. - Boundary markers: None identified; parameters are directly interpolated into script logic.
- Capability inventory: Local execution of Python and SageMath scripts, and the ability to trigger external CLI tools.
- Sanitization: No explicit sanitization or input validation for mathematical parameters is provided in the code snippets.
Audit Metadata