The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill downloads and executes an installation script from https://mise.run via a shell pipe (curl ... | sh). This is the official installation method for the mise runtime manager.
[COMMAND_EXECUTION]: The skill modifies the user's shell configuration file (~/.zshrc) by appending an activation command (eval "$(~/.local/bin/mise activate)"). This ensures the toolchain is automatically initialized in new shell sessions.
[COMMAND_EXECUTION]: The skill creates and modifies executable scripts within the .git/hooks directory (commit-msg, pre-commit, pre-push). These hooks are configured to execute mise tasks during the git workflow.
[COMMAND_EXECUTION]: Automates various development tasks including code generation, database migrations, and CI/CD pipelines using tools like mise, uv, bun, and flutter.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection.
Ingestion points: Parses untrusted data from git log and git diff outputs in resources/validation-pipeline.md and resources/release-coordination.md.
Boundary markers: No specific delimiters or instructions to ignore embedded commands are present.
Capability inventory: Includes the ability to execute shell commands via mise run, bunx, uv run, and docker compose across multiple scripts.
Sanitization: No sanitization or validation of the ingested git data is performed before it is used to determine execution logic.

oma-dev-workflow