The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute several shell commands as part of the documentation update workflow.
It uses grep to search for terms across the project files during non-accretive updates.
It attempts to execute a local project-specific script located at scripts/check-doc.sh to verify documentation integrity.
It performs version control operations using git add, git commit, and git push to finalize the session's work.
[DATA_EXFILTRATION]: The skill uses git push to transmit local documentation changes to a remote repository. While this involves sending data over the network, it is the intended and primary function of the skill within a developer's workflow.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes external "evidence" such as CI run URLs and command outputs to verify if tasks are completed. This introduces a surface for indirect prompt injection where malicious content in those external sources could attempt to influence the agent's behavior during the update process. However, the risk is inherent to the task of processing external logs and is mitigated by the agent's standard safety protocols.

wrap-up