The Agent Skills Directory

[SAFE]: The skill acts as a coordinator that routes user input to specific research sub-skills like echo-feedback and echo-interview based on arguments provided in {{args}}.
[SAFE]: Analysis of the instruction set found no evidence of prompt injection, data exfiltration, persistence mechanisms, or obfuscation. The permissions requested in the frontmatter align with the described user research capabilities.
[SAFE]: The skill processes untrusted user data as a surface for indirect prompt injection.
Ingestion points: User input enters the skill via the {{args}} variable in SKILL.md.
Boundary markers: No delimiters are used to wrap the interpolated {{args}} content.
Capability inventory: The skill has access to tools including Bash, Write, WebFetch, and WebSearch as defined in its allowed-tools metadata.
Sanitization: The user input in {{args}} is passed directly to the dispatch mechanism without visible filtering or sanitization.

echo