Skill Router Mod — Development skill for Claude Code
The skill-router mod: Jev picks and loads the skills each prompt needs.
How to install Skill Router Mod
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open gecm0/skill-router-mod and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Skill Router Mod does
The skill-router mod: Jev picks and loads the skills each prompt needs.
Alternatives in Development
- Maintainer — Meta-skill that loads all llmwiki governance docs and exposes the three maintainer slash commands 375 ★
- Paper Search Pro — Academic literature discovery as a Skill — Claude Code · Codex · any agent that loads SKILL.md 145 ★
- Inngest Functions — How to create dynamic Inngest functions that the sidecar loads automatically 135 ★
README
skill-router
A Claude Code mod that suggests the right skills for each prompt. On every prompt it asks [Jev](https://typesafe.ai) whether each skill fits, shows the ones over the threshold ranked above the prompt, and passes them to the model as a suggestion. It never orders a skill loaded: the model decides.
Jev suggests · tdd 0.96 · animate 0.95 (240 ms)
How it works
- Candidates are the skills the model can invoke itself (the engine's own listing), minus any the main conversation already loaded. User-only slash commands are left out, and a skill installed twice under two names is offered once.
- One Noul question per skill, all in one request: "is the user asking for this kind of task to be done now?" Each skill gets its own probability, so none at all is a normal outcome, and a prompt that only asks about or weighs a task does not trigger it.
- Skills at or over the threshold are ranked, up to the maximum, and passed to the model as a hidden note: a suggestion to load one only if it clearly helps, and never one that runs actions or delegates work unless you asked.
- Why only suggest: a mod cannot tell a skill that only teaches from one that acts (delegates to another agent, runs a subagent), and every loaded skill costs context.
- Skills the main conversation loads are remembered and not suggested again;
a subagent's loads do not count, and
/clearforgets them. - Fails open: a timeout, an outage or a missing key only skips the routing, and the band says why. The band clears when a prompt is not routed.
- Skips slash commands, prompts under 12 characters, and prompts not typed by you.
Options
Set in `/plugin` → Installed → skill-router → Configure options.
| Option | Default | Meaning |
|---|---|---|
| Threshold | 0.8 | Probability (above 0, up to 1) a skill needs to be suggested; anything else falls back to 0.8 |
| Max suggestions | 3 | Skills over the threshold to show and suggest, highest first |
| Timeout |
Related Skills
0x002132 Skills Bugbounty Hunt Dispatch
Skill-set loader for /hunt orchestrator. Fingerprints the target, picks the right platform attack skills, and
A2a Mod
A2A protocol client for Claude Code: Claude hands tasks to non-Claude worker agents over Agent2Agent and picks
Cronenberg
Meta-router: explicit /cronenberg picks the right pickle metaphor + cleanup chain for a build/implement reques
Jev Permission Gate
A Claude Code mod that uses TypeSafe's Jev to decide auto mode tool calls. 2x faster than the built-in classif
Carousel Cover
Generate a cover image for the current carousel via Higgsfield CLI. Loads the matching prompt template, fills
Jev Guard
Probability-scored guardrails for Claude Code: deny rule-breaking edits and unasked-for deploys, route your do
Related Agents
Typesafe Adversary
Red-teams TypeSafe integrations before users do. Probes the nine documented Jev failure modes, tests prompt in
Restaurant Router
Router agent for restaurant reservation requests. Reads the provider registry, picks the right provider agent
Ruver Fd Triage
Ruver FD router. Classifies goal as feature/bug/chore and picks full_feature vs debug_fix vs light_change. No