OpenAgentsControl/.opencode/context/core/system/context-guide.md

Context System Guide

Quick Reference

Golden Rule: Fetch context when needed, not before (lazy loading)

Key Principle: Reference files WITHOUT @ symbol - agent fetches only when needed

Structure: standards/ (quality + analysis), workflows/ (process + review), system/ (internals)

Session Location: .tmp/sessions/{timestamp}-{task-slug}/context.md

---

Overview

Context files provide guidelines and templates for specific tasks. Use lazy loading (fetch when needed) to keep prompts lean.

Available Context Files

All files are in .opencode/context/core/ with organized subfolders:

Standards (Quality Guidelines + Analysis)

• .opencode/context/core/standards/code.md - Modular, functional code principles
• .opencode/context/core/standards/docs.md - Documentation standards
• .opencode/context/core/standards/tests.md - Testing standards
• .opencode/context/core/standards/patterns.md - Core patterns (error handling, security)
• .opencode/context/core/standards/analysis.md - Analysis framework

Workflows (Process Templates + Review)

• .opencode/context/core/workflows/delegation.md - Delegation template
• .opencode/context/core/workflows/task-breakdown.md - Complex task breakdown
• .opencode/context/core/workflows/sessions.md - Session lifecycle
• .opencode/context/core/workflows/review.md - Code review guidelines

Lazy Loading (Recommended)

Reference files WITHOUT @ symbol - agent fetches only when needed:

"Write code following .opencode/context/core/standards/code.md"
"Review using .opencode/context/core/workflows/review.md"
"Break down task using .opencode/context/core/workflows/task-breakdown.md"

Benefits:

• No prompt bloat
• Fetch only what's relevant
• Faster for simple tasks
• Agent decides when to load

When to Use Each File

.opencode/context/core/standards/code.md

• Writing new code
• Modifying existing code
• Following modular/functional patterns
• Making architectural decisions

.opencode/context/core/standards/docs.md

• Writing README files
• Creating API documentation
• Adding code comments

.opencode/context/core/standards/tests.md

• Writing new tests
• Running test suites
• Debugging test failures

.opencode/context/core/standards/patterns.md

• Error handling
• Security patterns
• Common code patterns

.opencode/context/core/standards/analysis.md

• Analyzing codebase patterns
• Investigating bugs
• Evaluating architecture

.opencode/context/core/workflows/delegation.md

• Delegating to general agent
• Creating task context
• Multi-file coordination

.opencode/context/core/workflows/task-breakdown.md

• Tasks with 4+ files
• Estimated effort >60 minutes
• Complex dependencies

.opencode/context/core/workflows/sessions.md

• Session lifecycle
• Cleanup procedures
• Session isolation

.opencode/context/core/workflows/review.md

• Reviewing code
• Conducting code audits
• Providing PR feedback

Temporary Context (Session-Specific)

When delegating, create focused task context:

Location: .tmp/sessions/{timestamp}-{task-slug}/context.md

Structure:

# Task Context: {Task Name}

Session ID: {id}
Created: {timestamp}
Status: in_progress

## Current Request
{What user asked for}

## Requirements
- {requirement 1}
- {requirement 2}

## Decisions Made
- {decision 1}

## Files to Modify/Create
- {file 1} - {purpose}

## Static Context Available
- .opencode/context/core/standards/code.md
- .opencode/context/core/standards/tests.md

## Constraints/Notes
{Important context}

## Progress
- [ ] {task 1}
- [ ] {task 2}

---
**Instructions for Subagent:**
{Specific instructions}

Session Management

Session Structure

.tmp/sessions/{session-id}/
├── context.md          # Task context
├── notes.md            # Working notes
└── artifacts/          # Generated files

Session ID Format

{timestamp}-{random-4-chars} Example: 20250119-143022-a4f2

Cleanup

• Ask user before deleting session files
• Remove after task completion
• Keep if user wants to review

Best Practices

✅ Use lazy loading (no @ symbol) ✅ Fetch only relevant context ✅ Create temp context when delegating ✅ Clean up sessions after completion ✅ Reference specific sections when possible ✅ Keep temp context focused and concise

Golden Rule: Fetch context when needed, not before.