The Agent Skills Directory

[PROMPT_INJECTION]: The skill documents functionality for hooks such as handleRequest and preMap that process data from external HTTP requests and data pipelines. This represents a potential surface for indirect prompt injection if the processed data contains malicious instructions.
Ingestion points: HTTP request bodies/headers in handleRequest and record arrays in pipeline hooks (documented in SKILL.md).
Boundary markers: No specific boundary markers or instruction-ignoring warnings are defined in the provided instructional snippets.
Capability inventory: Scripts have access to the integrator-api module for platform resource management and sjcl for cryptographic operations.
Sanitization: The documentation does not explicitly mandate sanitization or validation patterns for incoming external data.
[COMMAND_EXECUTION]: The skill provides documentation for several celigo CLI commands used to manage the lifecycle of scripts, such as celigo scripts logs, celigo scripts create, and celigo scripts debug-enable. These are standard administrative tools for the Celigo ecosystem.

writing-scripts