decompose-into-slices
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple sources to generate its outputs.
- Ingestion points: Reads project context from
M###-CONTEXT.md, existing roadmap files, and external GitHub issues viamcp__github__issue_read. - Boundary markers: The instructions do not specify any delimiters or safety prompts to ignore instructions embedded within the ingested data.
- Capability inventory: The skill can spawn a sub-agent (
Agent(subagent_type=Explore)), write files to the.gsd/milestones/directory, and create GitHub issues usingmcp__github__issue_write. - Sanitization: There is no explicit sanitization or validation of the external content before it is used to draft slices or file issues.
- [EXTERNAL_DOWNLOADS]: The skill fetches data from a well-known service (GitHub) using the
mcp__github__issue_readtool to retrieve project information. - [COMMAND_EXECUTION]: The skill uses a sub-agent capability to explore the codebase. This is a controlled execution within the agent framework used to align proposed task slices with the existing code structure.
Audit Metadata