The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests a very large documentation surface (over 900 Markdown files) into the agent's context. A file reputation scanner identified references/docs/data-center/get-help/troubleshooting/troubleshooting-behaviours.md as malicious (FileRepMalware). Malicious content in the documentation surface can serve as a vector for indirect prompt injection, potentially influencing the agent's behavior.
Ingestion points: references/docs/.
Boundary markers: SKILL.md includes explicit instructions to treat reference content as product evidence and not as agent instructions.
Capability inventory: No executable code in the skill, but documentation includes many script examples with network and system access.
Sanitization: None identified for processed documentation content.
[PROMPT_INJECTION]: Static detectors flagged references/docs/connect/get-started/find-objects-ids-in-monday-com.md for containing patterns that attempt to enable unrestricted or dangerous modes, which could be used to bypass agent safety filters.
[NO_CODE]: The skill consists entirely of documentation and metadata; it does not ship with any internal executable scripts (Python, JavaScript, or Shell).

scriptrunner-documentation