State Management Skill

This skill provides techniques for managing workflow state through simplified, consolidated markdown files. The refactored structure uses just two file patterns: a single FLOW.md at project level and one ITEM-XXX.md per work item.

Core Principles

1. Minimal Footprint: Only two file patterns instead of 9+
2. Single Source of Truth: FLOW.md contains backlog + project context, ITEM-XXX.md contains everything about one item
3. Human Readable: All state in plain markdown, easy to edit
4. Checkpoint Recovery: State enables resumption from any interruption
5. Context Budget Awareness: Track and report context usage

Directory Structure

.flow/
├── FLOW.md           # Project state + backlog (single file)
└── items/
    ├── ITEM-001.md   # All state for work item 001
    ├── ITEM-002.md   # All state for work item 002
    └── ...

State Files Overview

File	Purpose	Contains
FLOW.md	Project-level state	Vision, backlog, active item, capabilities cache
ITEM-XXX.md	Per-item state	Phase, decisions, requirements, tasks, checkpoint

FLOW.md Structure

The single project file contains:

Sections

1. Vision: High-level project vision
2. Backlog: Table of all work items with status
3. Active Item: Currently active item pointer
4. Capabilities Cache: Discovered plugin mappings with timestamp

Template

See templates.md for the full FLOW.md template.

ITEM-XXX.md Structure

Each work item has a single consolidated file containing:

Sections

1. Header: Item ID, title, timestamps
2. Phase: Current phase and progress (DISCUSS/PLAN/EXECUTE/VERIFY)
3. Decisions: Numbered decisions with rationale
4. Requirements: Extracted functional/non-functional requirements
5. Tasks: XML-structured atomic tasks (when in EXECUTE)
6. Checkpoint: Current state snapshot for resume

Template

See templates.md for the full ITEM-XXX.md template.

Work Item States

Items progress through these phases:

Phase	Description	Next Phase
BACKLOG	Defined but not started	DISCUSS
DISCUSS	Gathering requirements	PLAN
PLAN	Creating task plan	EXECUTE
EXECUTE	Implementing tasks	VERIFY
VERIFY	Validating implementation	DONE
DONE	Completed	(terminal)
ON_HOLD	Paused intentionally	Previous
BLOCKED	Waiting on dependency	Previous

State Operations

Initialize Project

1. Create .flow/ directory if not exists
2. Create .flow/items/ directory
3. Scan for installed plugins
4. Detect project type from file patterns
5. Create FLOW.md with empty backlog + capabilities

Create Work Item (/flow-workflow:start)

1. Read FLOW.md to get next item number
2. Generate ITEM-XXX ID
3. Create .flow/items/ITEM-XXX.md with DISCUSS phase
4. Update FLOW.md backlog table
5. Set as active item in FLOW.md

Switch Active Item

1. Update checkpoint in current item's file
2. Update FLOW.md active item pointer
3. Load new item's context

Phase Transition

1. Verify completion criteria for current phase
2. Update phase in ITEM-XXX.md
3. Create checkpoint snapshot
4. Initialize next phase section

Create Checkpoint

1. Record current phase and task progress
2. Capture context budget percentage
3. List next recommended action
4. Add timestamp

Resume from Checkpoint

1. Read FLOW.md to get active item
2. Read ITEM-XXX.md checkpoint section
3. Restore context from checkpoint
4. Identify next action
5. Continue workflow

Context Budget Tracking

Track context usage in FLOW.md and report:

## Context Monitor

- Current session: [X]% (limit: 50% before fresh agent)
- Last auto-spawn: [TIMESTAMP]
- Fresh agents this session: [N]

State Validation

Before any phase transition, validate:

1. Completeness: Required sections filled
2. Consistency: No contradictory decisions
3. Recoverability: Checkpoint exists for current progress

Best Practices

Writing State Updates

• Use timestamps in ISO format
• Include brief rationale for decisions
• Mark status clearly (DONE, BLOCKED, etc.)
• Keep checkpoint current

Reading State for Context

• Load FLOW.md first for active item
• Load ITEM-XXX.md for current phase details
• Check checkpoint before proceeding

Handling Phase Transitions

• Always create checkpoint before transition
• Update both FLOW.md backlog and ITEM-XXX.md
• Announce transition to user

Migration from Old Format

If legacy .flow/ structure detected (BACKLOG.md, ACTIVE.md, PROJECT.md, etc.):

1. Read existing state files
2. Consolidate into FLOW.md
3. Consolidate per-item files into ITEM-XXX.md
4. Backup old files to .flow/legacy/
5. Report migration complete

Integration Points

State management integrates with:

• Capability Discovery: Cache plugin mappings in FLOW.md
• Workflow Orchestration: Phase transitions and checkpoints
• Smart Continuation: Resume from checkpoint via /flow-workflow:go

See templates.md for full file templates.