consulting-playbook
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection vulnerability surface through its core file-management and coaching workflow.
- Ingestion points: The skill ingests untrusted user data during the 'Mandatory Intake' (Q1-Q3) in
SKILL.mdand during the interactive coaching steps in all three sub-skills. - Boundary markers: The instructions lack explicit boundary markers or 'ignore embedded instructions' delimiters when writing user-derived content to the
blueprint/Blueprint-YYYY-MM-DD.mdfile. - Capability inventory: The agent has the capability to create directories, write/update markdown files in the
blueprint/path, and invoke sub-skills via the 'Skill tool'. - Sanitization: There is no mention of sanitizing, escaping, or validating the user's input before it is paraphrased and persisted in the blueprint file, which serves as continuous context for the agent's logic.
Audit Metadata