The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill operates as a central router that ingests user requests to determine the appropriate child skill for execution, creating a surface for processing untrusted data.
Ingestion points: User requests for code reviews, design audits, and strategic questioning are processed via SKILL.md and references/routing-guide.md.
Boundary markers: The skill does not define explicit delimiters to separate user input from its internal routing instructions.
Capability inventory: This router skill itself lacks high-risk capabilities such as network exfiltration, arbitrary command execution, or file system modification.
Sanitization: No specific input sanitization or filtering is implemented in the routing logic to handle potentially malicious payloads within user requests.
[EXTERNAL_DOWNLOADS]: The skill provides installation instructions that involve fetching components from the author's repository.
The README.md file suggests using a command to add the skill from the alvarovillalbaa/plugins repository, which belongs to the skill's vendor.

review