seo-flow

Warn

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The /seo flow sync command in SKILL.md executes a Python script located at sync_flow.py. This script is responsible for updating the skill's local prompt library and interacts with the GitHub CLI (gh).
  • [DYNAMIC_EXECUTION]: The skill utilizes dynamic execution to run sync_flow.py at runtime when triggered by the user to manage content updates. The logic for how this script handles data and network operations is not available for inspection within the provided files.
  • [EXTERNAL_DOWNLOADS]: The skill synchronization feature fetches markdown prompt files from the external repository github.com/AgriciDaniel/flow. Additionally, references/bibliography.md contains numerous links to external SEO industry websites used for research and data verification.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill accepts user-provided URLs and topics through commands such as /seo flow find [url|topic], /seo flow leverage [url], and /seo flow optimize [url].
  • Boundary markers: While the prompts contain instructions like "Use only the supplied inputs" and "Clearly label any assumption," there are no formal technical delimiters used to separate untrusted user-supplied data from the framework's instructions.
  • Capability inventory: The skill maintains the capability to execute shell scripts and perform network operations via the sync mechanism.
  • Sanitization: The skill does not describe any specific sanitization or filtering processes applied to external content before it is processed by the AI agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 24, 2026, 02:25 AM
Security Audit — agent-trust-hub — seo-flow