team-plan
Pass
Audited by Gen Agent Trust Hub on Nov 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple external files which could contain malicious instructions designed to influence the planning agent.
- Ingestion points: Reads
1-task.md,5-research.md,6-design.md, and7-structure.mdfrom the directory specified in$ARGUMENTS. - Boundary markers: None identified; instructions do not explicitly mandate delimiters or ignore-embedded-instruction warnings for the ingested content.
- Capability inventory: Executes a shell script (
discover-topic.sh), dispatches a specializedplanneragent, and writes output to8-plan.mdon the local file system. - Sanitization: No explicit sanitization or validation of the input file content is described before processing.
- [COMMAND_EXECUTION]: The skill executes a local discovery script using arguments provided at runtime, which is a standard but sensitive operation.
- Evidence:
SKILL.mdcontains a shell snippet:"<team-skill-dir>/discover-topic.sh" "${ARGUMENTS:-}" "7-structure.md". While the variables are double-quoted to prevent basic word splitting, the command relies on the integrity of the$ARGUMENTSvariable and thediscover-topic.shscript logic.
Audit Metadata