aeo-audit
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill accesses local files within the
~/.claude-marketing/directory, including brand profiles, manifest guidelines, and agency SOPs. While these are application-specific paths, reading data from the user home directory constitutes a surface for local data exposure. - [INDIRECT_PROMPT_INJECTION]: The skill processes brand data and external guidelines to configure its auditing behavior, creating a potential vector for indirect injection from those data sources. \n
- Ingestion points: Brand profiles (
profile.json), manifest guidelines (_manifest.json), and agency SOPs retrieved from~/.claude-marketing/(SKILL.md). \n - Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the ingested brand data as untrusted or to ignore embedded instructions. \n
- Capability inventory: The skill generates structured reports to the plugin data directory and references an internal script at
scripts/geo-tracker.py(SKILL.md). \n - Sanitization: The audit process does not describe any validation or sanitization steps for the content found in the ingested JSON files.
Audit Metadata