The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill determines the execution path for several Python scripts (validate_seo_findings.py, manage_markdown_site.py, manage_markdown_work.py) by reading a suite_root value from a local configuration file named .seo-suite-runtime.json located beside the skill definition. This dynamic loading from a computed path could allow an attacker to redirect the agent to execute malicious scripts if they can modify the configuration file.
[COMMAND_EXECUTION]: The instructions direct the agent to invoke the python interpreter to run scripts from the resolved <suite-root> directory for tasks such as validating findings and managing site implementation.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data from competitor websites and sources, which may contain malicious instructions intended to manipulate the agent.
Ingestion points: Competitor product pages, independently verifiable URLs, and raw data captures stored in the raw/ directory.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within external content are provided in the skill files.
Capability inventory: The skill possesses the capability to execute shell commands via Python and perform file system write operations.
Sanitization: There is no evidence of sanitization, filtering, or validation of the content retrieved from external competitor sources before it is utilized by the agent.

seo-competitor-pages