The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external artifacts, creating a surface for indirect prompt injection.
Ingestion points: The Workflow section instructs the agent to "Read the target artifact end to end."
Boundary markers: The instructions lack explicit requirements for using delimiters or boundary markers to distinguish between instructions and the data being processed.
Capability inventory: No external tools, shell commands, or network operations are requested or utilized by this skill.
Sanitization: The skill includes instructions for "edit-safety" and "unicode-safe" mutations, but these are focused on text integrity rather than sanitizing potential malicious instructions within the source text.
[NO_CODE]: The skill provides logic and rules for the AI agent to follow but does not ship with any accompanying scripts, binaries, or automated tooling configurations.

detool