seo-report-writer
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests raw analysis material from external sources such as audit findings and data-pull outputs, which provides an attack surface for indirect prompt injection. * Ingestion points: Analysis findings and data-pull outputs referenced in the Inputs section of SKILL.md. * Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore or isolate instructions embedded within the ingested data. * Capability inventory: The skill executes shell-level commands for scripts/project_memory.py, scripts/drift_store.py, and scripts/report_publish.py. * Sanitization: No data validation or sanitization routines are defined in the instructions.
- [COMMAND_EXECUTION]: The skill executes local Python scripts using parameters like domain names, subjects, and dates derived from potentially untrusted inputs. This creates a risk of command injection if the input values are maliciously crafted to include shell meta-characters.
- [EXTERNAL_DOWNLOADS]: The skill instructions specify adding a footer to every report that links to a non-whitelisted external personal website (https://leebeirne.com). While not a direct execution risk, it is an external reference to an untrusted domain.
Audit Metadata