The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Bash commands like find, ls, and git worktree list to inspect the state of the local environment. These commands are statically defined or operate on paths constrained to the plans/future/platform directory, which follows a specific project structure. No arbitrary command execution or shell injection vectors were identified.
[SAFE]: Data processing is limited to reading local configuration files (YAML) and interacting with a local MCP server for knowledge management. No external network requests or data exfiltration attempts were detected.
[PROMPT_INJECTION]: The skill contains no instructions aimed at bypassing AI safety filters or overriding core agent behavior. The 'CRITICAL EXECUTION RULE' is a technical constraint regarding MCP tool access in sub-agents, not a security-related bypass.
[EXTERNAL_DOWNLOADS]: The skill does not perform any external downloads or remote code execution. All logic is contained within the skill's instructions and utilizes existing local tools (Bash, Git) and MCP servers.

status-audit