The Agent Skills Directory

[COMMAND_EXECUTION]: The skill constructs shell commands using git and yq by interpolating identifiers (such as TICKET, EPIC, and STORY-ID) retrieved from local project files. If these files contain shell metacharacters, it could lead to arbitrary command execution when the agent runs the maintenance scripts.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it reads untrusted data from docs/sprint-status.yaml and docs/tech-specs/ to determine parameters for spawning new agent sessions. • Ingestion points: docs/sprint-status.yaml and docs/tech-specs/{TICKET}-tech-spec.md. • Boundary markers: No boundary markers or 'ignore' instructions are used when passing parsed content to the orchestrate task. • Capability inventory: The skill can execute shell commands, delete local files in .ralph-progress/, and initiate new AI agent tasks. • Sanitization: No validation or escaping is performed on the content parsed from external files before it is used in command strings or instructions.
[EXTERNAL_DOWNLOADS]: The skill references the official GitHub repository for yq and the Homebrew package manager as sources for required dependencies. These are well-known and trusted resources for software development.

bf-resume