The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the Bash tool to run diagnostic commands including npm run build, git status, git log, ls, find, and grep. These are executed to collect status and metric data from the development environment.\n- [INDIRECT_PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by reading and summarizing data from potentially untrusted local sources.\n
Ingestion points: The skill ingests text from git log (commit messages), npm run build (build logs), and grep results (TODO comments in website/src/) as defined in SKILL.md.\n
Boundary markers: Absent. The skill uses markdown templates for reporting but lacks specific delimiters or instructions to ignore commands that may be embedded within the summarized data.\n
Capability inventory: The skill has access to Bash, Read, Grep, and Glob tools, which allow for command execution and file system interaction.\n
Sanitization: Absent. External content is interpolated directly into the status briefing templates without filtering or validation.

liaison