parallel-plan
Pass
Audited by Gen Agent Trust Hub on Mar 28, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill instructs the agent to ingest and act upon content from files in the
agents/directory of the repository. This creates an indirect prompt injection surface where a malicious file in the repo could influence the orchestrator's logic or role assignment.\n - Ingestion points: Content is read from
agents/*.mdand stream report files.\n - Boundary markers: While the skill uses structured templates for communication, it lacks explicit instructions to ignore potentially malicious content within the agent definition files.\n
- Capability inventory: The skill is capable of coordinating multiple subagents and assigning specific workstreams based on the ingested roles.\n
- Sanitization: The skill does not describe any validation or sanitization of the content found in the
agents/*.mdfiles.\n- [NO_CODE]: The skill consists of Markdown instructions, configuration files, and templates, but does not include executable scripts or source code.
Audit Metadata