The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from repository files like manifests and templates to generate documentation, which represents an attack surface for indirect prompt injection. * Ingestion points: The agent reads various project files including package.json, issue templates, and commit history. * Boundary markers: The skill employs a strict plan-confirm-apply workflow, requiring the user to review a diff before any changes are made. * Capability inventory: The skill has the ability to Write to the file system and execute Bash commands. * Sanitization: A guardrail is included to identify and replace sensitive tokens or environment variables with placeholders in generated setup steps.
[COMMAND_EXECUTION]: The skill executes a local shell script provided in its directory to resolve and merge project configurations. * Evidence: Instructions in REFERENCE.md tell the agent to execute a script named resolve-config.sh using the shell.

write-contributing