lite-router

A task router that classifies work and sends it directly to Sonnet or Haiku agents, two model tiers, without first using the Opus planning agent.

In plain words
What is it for?
Use it with `/crew lite` or `/crew task --lite`; the project must first have CodeCrew's configuration and code-index files.
Why use it?
It provides a lower-cost task-routing mode and avoids the extra planning step handled by Opus.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/d3x293/code-crew/lite-router
Any agent
npx skills add d3x293/code-crew --skill lite-router
Clone the repo
git clone --depth 1 https://github.com/d3x293/code-crew

Made for: Claude Code, Codex.

Per session 43 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,844 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00043 $0.01844
Opus 5 $0.00022 $0.00922
Sonnet 5 $0.00009 $0.00369
Haiku 4.5 $0.00004 $0.00184

Measured 2d ago against content hash 5f24d8278221, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

lite-router scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/lite-router/SKILL.md · 189 lines

How it starts

The opening of the file, as written. The whole thing — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Lite Router - Task Router (No Opus)

Performs task classification and routing WITHOUT spawning the CEO (Opus) agent. All classification happens in the current session context (Sonnet-level), and agents are dispatched using only Sonnet and Haiku tiers.

When to Activate

  • User runs /crew lite "description"
  • User runs /crew task --lite "description" (backward compatible)

Prerequisites

These files MUST exist (run /crew init first):

  • .claude/crew-profile.md — project context
  • .claude/crew-team.json — agent configuration
  • .claude/crew-skills.json — skill catalog
  • .claude/crew-index.json — codebase index (Layer 1)
  • .claude/crew-symbols.json — symbol map (Layer 2)

If any are missing, tell the user to run /crew init first.

Lite Mode Advantage

In LITE mode:

  • 0 Opus calls — CEO is bypassed entirely
  • Classification is done by this skill (runs in user's Sonnet session)
  • vp-engineering (Sonnet) takes over planning/architecture duties that CEO would normally handle

Routing Algorithm

Step 1: Read Context

1. Read .claude/crew-profile.md for project context
2. Read .claude/crew-team.json for available agents
3. Read .claude/crew-index.json for codebase awareness
4. Check for .claude/.index-stale — if present, invoke index-updater for listed files, then delete the marker

Step 2: Classify Task

Task Type (pick one):

Type Signals
bug-fix "fix", "bug", "broken", "error", "crash", "not working", "issue"
feature "add", "implement", "create", "build", "new", "feature"
refactor "refactor", "clean up", "reorganize", "restructure", "improve"
docs "document", "readme", "comment", "explain", "docs"
test "test", "coverage", "spec", "TDD", "unit test", "e2e"
security "security", "vulnerability", "auth", "permission", "XSS", "injection"
devops "deploy", "docker", "CI/CD", "pipeline", "kubernetes", "build"
review "review", "check", "audit", "quality"
performance "slow", "optimize", "performance", "speed", "memory", "cache"

Read the full file on GitHub · 189 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 189 lines · 43 tokens per session scan A 5f24d8278221

Subscribe to this mod's changes

lite-router is a skill published in the GitHub repository d3x293/code-crew (13 stars, last pushed 4mo ago), licensed MIT. It adds 43 tokens to every session and 1,844 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.