The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text to perform plain-language rewrites, which establishes an ingestion surface for potential indirect injections.
Ingestion points: The skill takes arbitrary text input as defined in SKILL.md (Procedure section).
Boundary markers: None; instructions do not specify the use of delimiters for user-provided text.
Capability inventory: None; the skill does not request any tools via allowed-tools and contains no scripts or binary files.
Sanitization: No explicit sanitization or filtering logic is provided; the skill relies on the core model's safety guardrails.
[NO_CODE]: The skill contains no executable code, shell scripts, or binary assets. It is composed entirely of Markdown instructional files and configuration.
[SAFE]: The skill references reputable external documentation from the International Organization for Standardization (ISO), the International Plain Language Federation, and the European Commission to define its linguistic framework.

iso-24495