Route agent decisions through Hermes JEV Skills
Use Hermes JEV Skills when an operator wants fast, low-cost typed decisions for model routing, skill selection, memory filtering, search screening, mailbox triage, and browser or computer-use next actions.
npx skills add agentskillexchange/skills --skill route-agent-decisions-through-hermes-jev-skills
Hermes JEV Skills packages Jev-powered decision helpers as installable agent skills and a Hermes plugin, with support paths for Hermes, Claude Code, Codex, and other agents that can read skill files. Use it when an operator wants small repeated decisions moved out of frontier-model prose generation: choose a model tier, pick relevant memory passages, select installed skills, screen web results for prompt injection, classify inbox items, or choose the next safe GUI action from an approved table. Invoke this instead of handling every decision inside the main agent when the workflow needs measurable, typed, low-latency routing or filtering with fail-open behavior and privacy boundaries. The scope boundary is decision assistance for agent operations, not a generic Jev API card, a Hermes product listing, or a catch-all model-routing framework.
What this skill actually does
Inputs and prerequisites: Hermes JEV Skills, Python 3.9 or newer, TypeSafe/Jev API key for live Jev decisions, Hermes/Claude Code/Codex installation as applicable.
Setup notes: Clone https://github.com/kerpopule/hermes-jev-skills and run python3 install.py, then run jev setup-key and jev doctor. Use python3 install.py –check for a dry run before installing.
Source and verification boundary: use https://github.com/kerpopule/hermes-jev-skills as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.