The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it reads and summarizes external data sources.\n
Ingestion points: Reads activity data from the log/ directory and git log output.\n
Boundary markers: Absent. The instructions do not explicitly warn the agent to ignore potential instructions embedded in log files or commit messages.\n
Capability inventory: The skill utilizes the Edit tool to modify the .context/current-focus.md file and Bash for metadata retrieval.\n
Sanitization: Absent. Content retrieved from logs and git is formatted directly into the summary document without filtering.\n- [COMMAND_EXECUTION]: The skill is authorized to use the Bash tool for specific commands including git log, date, and ls. These commands are appropriately restricted to information gathering and do not pose a risk of arbitrary execution or privilege escalation.

update-focus