The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration attempts were detected in the skill instructions or associated scripts.
[COMMAND_EXECUTION]: The skill includes a utility script scripts/task-state.sh for managing task status transitions. The script implements safe shell scripting practices, including input validation via regular expressions for the owner field, atomic file operations using mktemp and mv, and concurrency control via flock to prevent race conditions during multi-agent coordination.
[EXTERNAL_DOWNLOADS]: The documentation references official, trusted sources for project configuration, such as fetching boilerplate .gitignore files from GitHub's official repository. It also defines strict policies for dependency management, requiring verification against official language registries (npm, PyPI, crates.io, etc.).
[SAFE]: Process management instructions for tmux and local routing via nsl follow standard developer practices and do not request excessive privileges or perform suspicious background operations.

pma