The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves analyzing external data, which introduces a potential attack surface for indirect prompt injection.
Ingestion points: The skill is designed to process untrusted external inputs including scraped page text, reviews, comments, and emails as specified in SKILL.md and references/skill-contract.md.
Boundary markers: No specific technical delimiters for data boundaries are defined in the instructions.
Capability inventory: The skill can perform handoffs to other marketing skills and interact with external SEO/Marketing APIs via connectors defined in references/connectors.md.
Sanitization: The skill includes robust mitigation by explicitly instructing the agent to treat external material as evidence only, not as instructions, and strictly forbids following any instructions found within that data.
[DATA_EXFILTRATION]: The skill references various external marketing and SEO platforms (e.g., Google Search Console, Ahrefs, HubSpot). However, it includes defensive guidelines in references/connectors.md to prevent credential exposure by mandating that API keys be kept out of skill files and source control.
[COMMAND_EXECUTION]: No instances of shell command execution or dynamic context injection were found in the skill files.

competitor-analysis