money-content
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions direct the agent to ingest and analyze untrusted external data, including codebase contents, documentation, competitor websites, and project changelogs to inform content generation.
- Ingestion points: Stage 1 Research (codebase, landing page, docs, competitor content), Release-Notes Mode (CHANGELOG.md, deploy log, learnings.jsonl).
- Boundary markers: The skill does not define explicit delimiters or specific instructions for the agent to ignore potential malicious content or instructions embedded within the ingested data.
- Capability inventory: The agent is expected to use file-reading and web-browsing capabilities to process these inputs across all research and generation stages.
- Sanitization: No sanitization, validation, or filtering steps are defined for the external content before it is interpolated into prompts for generating marketing artifacts.
- [DATA_EXPOSURE]: The skill accesses local project metadata and historical learning files located in the
~/.smtm/projects/directory. These files (e.g.,profile.json,learnings.jsonl) contain business profiles and conversion data necessary for the skill's stated purpose of automated growth strategy.
Audit Metadata