deal-playbook
Pass
Audited by Gen Agent Trust Hub on Jun 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions focus on transforming user-provided business context into sales playbook artifacts. There are no instructions that would cause the agent to execute arbitrary code or exfiltrate sensitive information.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it ingests untrusted data from the user to generate content.
- Ingestion points: User-provided context in Step 1 (Target role, Trigger event, Primary pain, etc.).
- Boundary markers: Absent; the input is directly used to populate variables in the playbook templates.
- Capability inventory: The skill specifies 'Bash', 'WebFetch', and 'WebSearch' in its allowed-tools, though they are not explicitly invoked for processing the untrusted input in the instructions.
- Sanitization: Absent; no validation or escaping of user input is specified before generation. This is a low-risk surface common to creative writing agents.
- [DATA_EXPOSURE]: No hardcoded credentials, API keys, or sensitive file paths (e.g., .ssh, .env) were found in the skill or its configuration.
- [EXTERNAL_DOWNLOADS]: The skill does not download external scripts or executable binaries. It references a local document 'docs/output-kit.md' for formatting guidance, which is a safe internal reference.
Audit Metadata