data
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill facilitates standard binary analysis tasks without any evidence of malicious intent, obfuscation, or unauthorized network access. All provided SQL examples and documentation align with legitimate reverse engineering workflows.\n- [PROMPT_INJECTION]: Analysis identified a surface for indirect prompt injection because the skill allows the agent to read raw strings and bytes from potentially untrusted binaries. 1. Ingestion points: The 'strings' table and the 'bytes()', 'bytes_raw()', and 'search_bytes()' SQL functions in SKILL.md. 2. Boundary markers: No specific delimiters or safety instructions are used to distinguish ingested binary data from system prompts. 3. Capability inventory: The agent has access to Bash, Read, Glob, and Grep tools as specified in SKILL.md. 4. Sanitization: No sanitization of the ingested binary content is described. This surface is inherent to the tool's primary purpose and does not represent a malicious finding in the skill itself.
Audit Metadata