The Agent Skills Directory

[SAFE]: The skill serves as a guide for security reporting and vulnerability triage. All operations described, such as CVSS scoring and root-cause analysis, align with the tool's primary purpose.
[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted external data such as source code, binaries, and network traces, identifying a surface for indirect prompt injection. • Ingestion points: Static code, binary analysis, and request/response traces (SKILL.md). • Boundary markers: No explicit delimiters or instructions to ignore embedded instructions are present. • Capability inventory: Mentions file searching, binary inspection, and test execution (SKILL.md). • Sanitization: No specific sanitization or validation of the processed data is mentioned.

offensive-reporting