The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes 'plans or instructions' which are potentially untrusted inputs. * Ingestion points: The instruction 'Implement the plan/instructions' in step 2 (SKILL.md) indicates data entry from the environment or user. * Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present. * Capability inventory: The skill can execute shell commands via pnpm qa and perform network operations via git push. * Sanitization: No evidence of validation or sanitization of the input instructions before execution.
[COMMAND_EXECUTION]: The skill uses shell commands for project management and testing. * Evidence: The skill executes git pull, git checkout, pnpm qa, and git push to perform its primary functions.

exec-push