multi-thread-orchestration
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external information provided by the user regarding account stakeholders, buying committees, and deal dynamics.
- Ingestion points: User-supplied deal data, stakeholder mapping details, and "ghost node" identifiers are incorporated into the agent's context (found in SKILL.md).
- Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the ingested data.
- Capability inventory: The skill utilizes a local Python script (
scripts/check-output.py) for file reading and deliverable validation. - Sanitization: No explicit sanitization or filtering of external content is specified before processing or script execution.
- [DYNAMIC_EXECUTION]: The skill provides and instructs the agent to execute a local Python utility script for output validation.
- Evidence:
scripts/check-output.pyis included in the skill to verify that generated deliverables meet minimum criteria. The agent is directed to run this script as part of its workflow (referenced in SKILL.md and references/framework-notes.md). - Execution method: Local Python subprocess call. The script reads a file path provided as a command-line argument and performs basic keyword checks.
Audit Metadata