The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute the bandit command-line tool to analyze Python files. This is a legitimate and expected use of shell capabilities for a security scanning task.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted Python source code and parses the resulting scan output, which constitutes a potential surface for indirect prompt injection. 1. Ingestion points: Local Python files and bandit-results.json (SKILL.md). 2. Boundary markers: No specific boundary markers or warnings to ignore instructions within the code are defined. 3. Capability inventory: Execution of the bandit tool via shell (SKILL.md). 4. Sanitization: No explicit sanitization or validation of the input code is described in the workflow.

sast-bandit