The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external project data (Crystal source code) through the Ameba linter, representing a potential surface for indirect instructions if the analyzed code is malicious.
Ingestion points: Reads and analyzes local .cr files via git diff and ameba commands.
Boundary markers: The scripts do not implement explicit boundary delimiters or warnings against embedded instructions in the source code.
Capability inventory: The skill utilizes Bash for command execution and Read for file access.
Sanitization: No specific sanitization or filtering is performed on the files before analysis.
[DYNAMIC_EXECUTION]: The skill facilitates the creation and execution of local shell scripts to automate Git hooks.
Evidence: Instructions describe creating shell scripts in .git/hooks/ and applying chmod +x to enable execution.
[COMMAND_EXECUTION]: Orchestrates the use of standard development tools including git, xargs, grep, and jq to manage linting results and workflows.
[EXTERNAL_DOWNLOADS]: Configures CI/CD pipelines to download official Crystal language tools and community-standard GitHub Actions.
Evidence: References official components such as crystal-lang/install-crystal and crystal-ameba/github-action for environment setup and reporting.

ameba-integration