feature-launch-playbook
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a template-driven reasoning prompt. It does not include executable code, third-party dependencies, or network-enabled commands. All operations are limited to text processing and writing output to the local filesystem.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external product specifications and PRDs to generate copy. While this creates a surface for processing untrusted data, the risk is minimal as the output is restricted to marketing text and markdown files. 1. Ingestion points: Phase 0 Intake (Notion docs, PRD excerpts, or plain text descriptions). 2. Boundary markers: Absent; the skill does not instruct the agent to use specific delimiters or ignore embedded instructions within the source text. 3. Capability inventory: File writing (Phase 4) to the local working directory. 4. Sanitization: Absent; no instructions provided to escape or validate the contents of the ingested feature context before asset generation.
Audit Metadata