The Agent Skills Directory

[SAFE]: The skill provides instructional guidance for developing ClickUp SDKs, specifically emphasizing security best practices such as avoiding credential serialization in logs and using governed secret references for authentication tokens.
[SAFE]: All external resource links point to official ClickUp developer documentation (developer.clickup.com) or their official GitHub repository (github.com/clickup), which are trusted/well-known services.
[INDIRECT_PROMPT_INJECTION]: The skill has a potential ingestion surface as it reads external documentation and API contracts using WebFetch.
Ingestion points: Reads official ClickUp documentation and OpenAPI schemas via WebFetch in SKILL.md and references/official-docs.md.
Boundary markers: The instructions explicitly specify redaction policies for telemetry and evidence, advising against including task text, comments, or credentials in output.
Capability inventory: The skill uses Read, Glob, Grep, Write, and Edit for repository management, but limits their use to specific boundaries mentioned in 'Tool Discipline'.
Sanitization: It mandates the redaction of sensitive identifiers, task text, and credentials from any generated evidence or receipts.

clickup-sdk-patterns