The Agent Skills Directory

[COMMAND_EXECUTION]: The skill constructs and executes shell commands such as git checkout -b 'BRANCHNAME' using variables derived from external sources (SKILL.md).
[PROMPT_INJECTION]: The skill processes data from external and local sources that could contain malicious instructions or unexpected characters.
Ingestion points: Data enters the context via the Linear API (ticket fields) and local files including .claude/commands/linear.md and .claude/commands/describe_pr.md (SKILL.md).
Boundary markers: There are no boundary markers or delimiters defined to isolate untrusted data or warn the agent to ignore embedded instructions.
Capability inventory: The agent has access to git and gh CLI tools, providing the ability to modify the project structure, commit code, and push to remote repositories.
Sanitization: Input variables like 'BRANCHNAME' and 'COMMITHASH' are not sanitized or validated before being passed to the shell, allowing for potential command injection if the external source is compromised.

founder-mode