workflow-quarterly-review
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill makes extensive use of command-line execution to run local Python scripts including
scripts/project_memory.py,scripts/drift_store.py,scripts/google_client.py,scripts/dfs_client.py,scripts/cost_ledger.py,scripts/report_build.py, andscripts/report_publish.pyto process SEO data and generate reports. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially contain malicious instructions aimed at the agent.
- Ingestion points: Data is pulled from Google Search Console (queries), Google Analytics 4 (organic trends), and web mentions (via
dfs_client.py mentions --keyword "<brand>") in SKILL.md. - Boundary markers: The instructions lack explicit boundary markers or delimiters to separate external data from system instructions.
- Capability inventory: The skill has the capability to execute multiple local subprocesses (Python scripts) and perform file system writes (generating
.mdand.htmlreport files). - Sanitization: There is no mention of sanitization, filtering, or validation of the ingested SEO data before it is interpolated into the narrative wins/losses summaries.
Audit Metadata