skill-router
Pass
Audited by Gen Agent Trust Hub on Jul 8, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of a local Python tool (
scripts/route_skill.py). This script performs string-based matching between user queries and a skills catalog (cli/skills.json). The script uses standard libraries and operates within the local file system without performing high-risk actions like network connections or persistent system modifications. - [PROMPT_INJECTION]: The skill processes user-provided goals as command-line arguments. To mitigate risks associated with untrusted input, the instructions specify wrapping the input in double quotes. Furthermore, the routing script internally sanitizes the input using the
tokenizeandnormalizefunctions, which strip non-alphanumeric characters before performing any matching logic. - [DATA_EXPOSURE]: The script reads from a local catalog file (
cli/skills.json) to provide recommendations. This file contains skill names, descriptions, and tags, which is consistent with the skill's stated purpose and does not involve accessing sensitive user credentials or system configuration files.
Audit Metadata