The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to execute local Python scripts, including validate_seo_action_plan.py, validate_seo_findings.py, manage_markdown_site.py, and manage_markdown_work.py. These scripts perform integrity checks on SEO artifacts and manage the markdown delivery workflow. The execution environment is determined by the CLAUDE_PLUGIN_ROOT or configurations within .seo-suite-runtime.json.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data bundles which could potentially contain adversarial content designed to influence the agent's output.
Ingestion points: The agent reads optimization-brief.json, research-pack.json, and seo-findings.json from the inputs/ directory, as described in SKILL.md and references/action-plan-protocol.md.
Boundary markers: The instructions do not specify explicit delimiters for the input data within the agent's context, although they mandate the use of validation scripts to confirm artifact structure.
Capability inventory: The skill has the capability to run Python scripts via the shell to validate plans and manage markdown site exports.
Sanitization: The skill relies on external validation scripts (validate-plan, validate-findings) to ensure the integrity and traceability of input bundles before processing.

seo-action-plan