The Agent Skills Directory

[COMMAND_EXECUTION]: The skill defines a /seo flow sync command intended to execute a Python script (scripts/sync_flow.py) for updating the framework's prompt library. It should be noted that the target script is not present in the provided file package, suggesting a dependency on external or environment-provided code.
[EXTERNAL_DOWNLOADS]: The framework is designed to synchronize and pull content from the author's GitHub repository (github.com/AgriciDaniel/flow). This represents a standard update mechanism leveraging the vendor's own infrastructure.
[INDIRECT_PROMPT_INJECTION]: The skill processes external URLs and topical data through its prompt library, creating a potential surface for indirect prompt injection where malicious instructions in audited content could attempt to influence the agent.
Ingestion points: External data enters the agent context via the [url|topic] arguments in the stage-specific commands (find, leverage, optimize, win, local) defined in SKILL.md.
Boundary markers: The prompt templates include explicit constraints instructing the agent to "Use only the supplied inputs" and "clearly label any assumption," which serve as behavioral controls against embedded instructions.
Capability inventory: The skill possesses the capability to write audit summaries to a local data cache (.seo-cache/) and to execute the synchronization script.
Sanitization: The skill relies primarily on model-level instructions within the prompts rather than explicit programmatic validation or sanitization of the external content.

seo-flow