sdlc-create-plan
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository content, including specifications and project files, to generate its implementation plans. This creates a surface for indirect prompt injection where malicious instructions embedded in a project's documentation could attempt to influence the agent's task generation.
- Ingestion points: Found in
SKILL.mdandnew/SKILL.md, which instruct the agent to inspectintent.md,spec.md, the repository, and current implementation. - Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious directions within the ingested artifacts.
- Capability inventory: The skill performs filesystem operations to create and update
plan.mdfiles and uses theAskUserQuestiontool to interact with the user. No high-risk capabilities like network access or arbitrary code execution were identified. - Sanitization: No sanitization or validation of the input markdown content is performed before it is used to generate the plan.
Audit Metadata