claude

model-routing

Route sub-agent (leaf) tasks to the right model by task type — measured twice on a real app build. Ships nine leaf-* agents (closed/executor work → low tier; open problems, discovery and the pre-merge gate → mid tier) that auto-delegate by description, plus an /apply command that adds the routing gate to your CLAUDE.md. Stops every delegation from silently burning your priciest tier — after the second pass no agent pins a top-tier model outright; a one-shot emergent implementation is the one place that tier is a measured win, and diagnose + repair keeps it too as an untested hedge for the hardest cases.

1 on GitHubby Jooeon Kimv2.0.0MITupdated 2 months ago
/

Plugin model-routing from marketplace agent-harness in K021/agent-harness. Commands read from the repo's manifest — 1 month ago.

What it installs

Slash commands (1)

  • /model-routing:apply

    Add the leaf model-routing gate to your CLAUDE.md and tailor it to your model lineup.

Subagents (9)

  • leaf-adversarial-gate

    The final pre-merge gate for emergent implementations and diagnosis-repairs — open-ended deep review that hunts for defects and blocks the merge. NOT for checklist-driven review (use leaf-checklist-review). This is the one role every "a cheaper model is fine" decision depends on, so don't cheap out

  • leaf-checklist-review

    Review where WHAT to look at is already specified as a checklist / set of lenses — pre-merge gate reviews, security/privacy review against given criteria, checking a doc against a requirements list. NOT for open-ended review where you must decide what to inspect; use leaf-adversarial-gate for that.

  • leaf-coverage-research

    Coverage research where the scope to collect is DEFINED (official docs, a fixed source list) and the goal is to gather completely and compare against a baseline — compatibility matrices, standard cross-checks, exhaustive collection. If the scope is open and "finding hidden things" is the point, use

  • leaf-design-mockup

    Build design mockups / comparison pages — several variants side by side for a human to pick (new screen or component drafts). The final choice is the human's. (Re-measured 2026-07: on both mockup tasks the mid tier was the only model to beat the previous best — and in the earlier round no single mod

  • leaf-diagnose-repair

    Diagnose AND fix a bug whose cause is NOT yet established (hypothesis-level) — diagnosis through repair in one go. If the fix method is already given, use leaf-oracle-repair. The output must pass leaf-adversarial-gate before merge. Re-measured 2026-07: this tier now **ties the top tier on defects (1

  • leaf-discovery-research

    Discovery research where finding what's OUTSIDE the official list — hidden features, undocumented constraints, workarounds, pitfalls — is the substance. For defined exhaustive collection use leaf-coverage-research. Discovery is the one axis where models genuinely split, and a missed finding can't be

  • leaf-emergent-impl

    Implementation where only the direction / cause-family is given and "the design is your call" — typical when several subsystems' rules (sync, concurrency, data integrity) must all hold at once. The output MUST pass through leaf-adversarial-gate before merge. For concurrency / data-integrity / securi

  • leaf-oracle-repair

    Fix or refactor where BOTH the method (what to change and how) AND an oracle (tests / output comparison that mechanically judges success) are supplied. Behavior-preserving refactors, bugs whose cause is already agreed. If the cause is unknown, use leaf-diagnose-repair. (First pass: with method + ora

  • leaf-plan-draft

    Produce a SCOPED PLAN document (not code): the goal is known and the work is detailing it — investigation and trade-off analysis for a restructure/migration. Output is always a "pre-review draft". If the goal itself is unknown and the work is searching broadly for what to do (an exploratory plan), o

View source on GitHub

More from K021/agent-harness

All 3 plugins →

Gives your AI coding agent a memory of the work — a self-maintaining context index (project-context.json) that survives session resets and context-window compaction, so the agent recovers what it's building, where it is, and the next step instead of re-deriving it.

1 skill · hooks

1

Gives your AI coding agent a map of the files — a hot index of what files exist and what each does (project-structure.json + append-only history), recovered in a single read so a new session stops re-discovering the codebase by grep. Sibling to project-context.

1 skill · hooks

1

Similar plugins

Matt Pocock's agent skills for real engineering (grilling, spec/ticket flows, TDD, code review, domain modelling and more).

39 skills

mattpocock/skills 323,108

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

mem0

claudecodex

Cross-session memory and token savings for coding agents.

6 skills · 1 agent · 1 MCP server · hooks

mem0ai/mem0 68,893

context7

claudecodex

Up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your coding agent's context.

1 MCP server

upstash/context7 64,709

Automates browser interactions for web testing, form filling, screenshots, and data extraction

12 skills

vercel-labs/agent-browser 50,373

pensyve

claude

Universal memory runtime for Claude Code with cross-session memory, entity-aware recall, lifecycle hooks, skills, commands, and memory-curator agents

7 skills · 7 commands · 2 agents · hooks

wshobson/agents 42,461

Maintain model-routing?

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.

claude & codex plugins3 plugins