| name | context-partitioning |
| category | context |
| version | 2.0.0 |
| description | Manifest-first context isolation — each subagent receives only its scope, never the full codebase |
| author | Unite Group (ported from NodeJS-Starter-V1) |
| priority | 1 |
| auto_load | true |
| triggers | ["any_dispatch","spawn_subagent","agent_harness"] |
| context | fork |
Context Partitioning — Manifest-First Dispatch
The Problem
Without explicit context isolation, the main orchestrator context window accumulates:
- Full file reads that were only needed once
- Subagent outputs that are already committed to disk
- Skill files that don't apply to the current task
- Historical reasoning that compaction will destroy anyway
Result: context bloat → compaction → rule violations → wrong output.
The Solution: Manifest-First
Rule: Before dispatching ANY subagent, the orchestrator MUST produce a typed manifest.
The manifest is the subagent's entire world — it reads NOTHING outside its manifest.
Manifest Format
{
"agent": "frontend-specialist",
"task": "Add persona avatar upload to SocialPersonasManager",
"toolshed": "frontend",
"token_budget": 60000,
"files": {
"must_read": [
"src/components/social/SocialPersonasManager.tsx",
"src/lib/businesses.ts"
],
"reference_only": [
"src/app/globals.css"
],
"must_not_touch": [
"src/lib/supabase/"
