sequence-writer
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of Markdown documentation and procedural instructions. There are no scripts (Python, JavaScript, Shell) or executable files included in the package.\n- [SAFE]: No malicious patterns or security risks were detected. The skill provides frameworks for cold outreach, including best practices for infrastructure setup (DMARC/SPF/DKIM) and content generation strategies.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied data such as outreach requests and account signals to generate copy.\n
- Ingestion points: The agent ingests user requests and account list data as described in
SKILL.md(Step 0, Step 2, and Step 3).\n - Boundary markers: The
SKILL.mdfile contains a 'Hard rules' section specifically designed to override user instructions and maintain behavioral boundaries.\n - Capability inventory: The skill does not define or utilize any capabilities for filesystem access, network communication, or shell command execution.\n
- Sanitization: No explicit sanitization of input data is defined, but the lack of executable capabilities mitigates the risk associated with untrusted data ingestion.
Audit Metadata