workflow-content-refresh
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches external web content via
webfetchto perform quality and overlap scans in Step 3. This ingested data could contain malicious instructions designed to influence the agent's behavior during the triage or briefing phases. - Ingestion points: The skill fetches candidate URLs using
webfetchas described inSKILL.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded content are defined.
- Capability inventory: The skill has the ability to write files to the local file system (SEO reports directory) and execute local Python scripts (
dfs_client.py,google_client.py,report_publish.py). - Sanitization: There is no mention of sanitization, filtering, or validation of the content fetched from external URLs before it is processed by the agent.
- [COMMAND_EXECUTION]: The workflow involves executing several local Python scripts to interact with ranking APIs, Google Search Console, and to generate reports.
- Evidence: Use of
python scripts/dfs_client.py,python scripts/google_client.py, andpython scripts/report_publish.pythroughout the workflow steps.
Audit Metadata