email-sequence
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill instructions in
SKILL.mdcreate a surface for indirect prompt injection by directing the agent to read and follow context from an external file (.claude/product-marketing-context.md). If this file were modified by a malicious actor (e.g., via a compromised repository or untrusted PR), it could influence the agent's behavior. - Ingestion points: The 'Initial Assessment' section of
SKILL.mdinstructs the agent to read.claude/product-marketing-context.mdbefore proceeding with the task. - Boundary markers: Absent. The instructions do not specify any delimiters (such as XML tags or markdown blocks) or provide 'ignore embedded instructions' warnings for the data ingested from the context file.
- Capability inventory: The skill produces text-based artifacts (email drafts, metrics plans, and architecture docs) and includes a local Python analysis script (
scripts/sequence_analyzer.py). - Sanitization: Absent. There are no instructions for the agent to validate, sanitize, or filter the content of the ingested context file.
Audit Metadata