house-rules
claudeEnforces aj's global CLAUDE.md rules as hooks: no hidden background work, no unasked git mutations, no unchecked destructive commands.
2 skills · 6 commands · 5 agents · hooks
STATUS: shell/v0.1. Classifies each prompt's complexity and nudges Claude to delegate to a model-pinned subagent - haiku for simple doc/comms work, sonnet for recon and code implementation, opus for planning, architecture, and management. Cannot switch the live session's own model.
Plugin agent-router from marketplace aj-house-rules in Ajw2003/AjsClaudeCodeTools. Commands read from the repo's manifest — 3 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
/agent-router:routeClassify the user's last request (or pasted text) into a routing tier and offer to delegate to the matching subagent
architectPlanning, architecture, and management-shaped work - system design, weighing tradeoffs, roadmaps, cross-cutting migrations, breaking a large ask into a plan, deciding what to delegate where. Use PROACTIVELY when agent-router's `route` hook flags a prompt as planning/architecture-tier and the current
operativeRecon and implementation - searching or reading a codebase to understand it, writing or fixing code against a plan that's already decided, debugging, running tests. Use PROACTIVELY when agent-router's `route` hook flags a prompt as recon/implementation-tier, so the work runs on Sonnet instead of Opu
scribeWrites or edits prose where the content is already decided - README updates, changelog entries, commit messages, code comments, docstrings, formatting fixes, straightforward renames. Use PROACTIVELY when agent-router's `route` hook flags a prompt as doc/comms-tier, so the work runs on Haiku instead
sh "${CLAUDE_PLUGIN_ROOT}/scripts/run.sh" inject
sh "${CLAUDE_PLUGIN_ROOT}/scripts/run.sh" route
Enforces aj's global CLAUDE.md rules as hooks: no hidden background work, no unasked git mutations, no unchecked destructive commands.
2 skills · 6 commands · 5 agents · hooks
STATUS: shell/v0.1. Notices an under-specified prompt and walks the user through goal, constraints, success criteria and gating before Claude runs with an inferred reading of it.
1 command · hooks
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.