The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by instructing the agent to ingest data from untrusted sources.\n
Ingestion points: The skill directs the agent to read a local marketing-context.md file and fetch robots.txt from user-defined external domains as part of the audit process.\n
Boundary markers: The instructions do not define boundary markers or specify that the ingested content should be treated as data rather than instructions.\n
Capability inventory: The skill is composed entirely of Markdown documentation and does not utilize any scripts, shell commands, or subprocess operations.\n
Sanitization: No sanitization or validation protocols are specified for the ingested content.\n- [METADATA_POISONING]: There is a mismatch between the author name in the SKILL.md frontmatter ('Alireza Rezvani') and the author identity provided in the system context ('UnlimitedxIQ'). Misleading metadata can be a technique used to misrepresent the origin of a skill.\n- [NO_CODE]: The skill does not contain any executable scripts, binaries, or source code for compilation. All files are informational Markdown documents, which eliminates the risk of direct remote code execution or persistence mechanisms within the skill package.

ai-seo