The Agent Skills Directory

[SAFE]: The skill serves as a reference guide for Keboola project structures and configuration files. It follows security best practices by explicitly advising users not to commit sensitive files like .env.local which contain API tokens.
[INDIRECT_PROMPT_INJECTION]: The skill describes a workflow involving the analysis and editing of external configuration files (config.json, meta.json, manifest.json). This presents a potential surface for indirect prompt injection if those files contain untrusted content.
Ingestion points: Local project configuration files including config.json and meta.json (SKILL.md).
Boundary markers: None explicitly defined to distinguish between configuration data and potential embedded instructions.
Capability inventory: The skill context implies the agent will perform file read/write operations and use CLI tools like kbc diff and kbc validate (SKILL.md).
Sanitization: No specific sanitization or validation protocols for the content of processed configuration files are mentioned.

keboola-config