The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it uses data provided by the user (such as organization slugs and repository names) directly in shell commands executed by the agent. \n
Ingestion points: Input collected via AskUserQuestion for organization slugs, repository names, and repository IDs in SKILL.md. \n
Boundary markers: No delimiters or instructions to ignore embedded commands are present when interpolating user data into command strings. \n
Capability inventory: The skill utilizes circleci CLI commands, git commands, and gh API calls to perform its tasks in SKILL.md. \n
Sanitization: The instructions lack specific validation or sanitization steps to ensure that user-provided strings do not contain malicious shell metacharacters. \n- [COMMAND_EXECUTION]: The skill executes various system commands using circleci, git, and gh to configure the environment and verify the setup. These operations are intended for the skill's primary function but rely on the integrity of the variables supplied during the workflow.

onboarding