task-decomposer
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data that could contain malicious instructions designed to influence the agent's behavior during task execution.\n
- Ingestion points: User-supplied task descriptions and domain-specific
ECOSYSTEM.mdfiles (referenced inSKILL.md) are read into the agent's context.\n - Boundary markers: The skill instructions do not specify the use of delimiters or "ignore" instructions to prevent the agent from following commands embedded in the task data or domain files.\n
- Capability inventory: The skill directs the agent to perform actions such as writing code and creating files, providing a pathway for injected instructions to result in tangible side effects.\n
- Sanitization: There is no mention of filtering, escaping, or validating the content of the ingested files or user prompts before they are incorporated into the subtask plans.
Audit Metadata