writing-plans
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted specifications to generate implementation plans containing shell commands and source code for downstream execution. A malicious input specification could influence the agent to generate harmful commands within the plan.
- Ingestion points: User-provided requirements or specifications (SKILL.md).
- Boundary markers: None are defined to isolate the specification content.
- Capability inventory: The generated plans include file creation, modification, and shell command execution such as git and pytest (SKILL.md).
- Sanitization: No explicit sanitization or validation of the generated plan against the input specification is performed.
- [DYNAMIC_EXECUTION]: The skill explicitly generates executable code and shell commands based on external input, which are intended to be run by subsequent tools.
- Evidence: The Task Structure section in SKILL.md provides templates for shell commands and Python code blocks that the agent is instructed to populate and execute.
Audit Metadata