Context Engineering

This skill should be used when designing context management, implementing tiered fidelity, reducing token waste, applying Four Laws patterns, creating "NOT PASSED" sections, optimizing agent context, or debugging context-related issues. Provides SOTA patterns for context-efficient multi-agent systems achieving 60-80% token reduction.

AI Agents

What this skill does


# Context Engineering

## Overview

State-of-the-art patterns for managing context in LLM agent systems. These patterns enable complex multi-agent workflows while minimizing token overhead through strategic context engineering.

## The Four Laws of Context Management

| Law | Principle | Token Impact |
|-----|-----------|--------------|
| **1. Selective Projection** | Pass only fields each agent needs | -30-50% |
| **2. Tiered Fidelity** | Define explicit context tiers per role | -40-60% |
| **3. Reference vs Embedding** | Use references for large data | -50-80% |
| **4. Lazy Loading** | Load data on-demand, not upfront | -30-50% |

For detailed explanations and examples, see `references/four-laws.md`.

## Context Tiers

| Tier | Description | Use Case | Typical Size |
|------|-------------|----------|--------------|
| **FULL** | Complete data | Initial analysis | 5-20K tokens |
| **SELECTIVE** | Relevant subset | Domain workers | 1-5K tokens |
| **FILTERED** | Criteria-matched | Validators | 500-2K tokens |
| **MINIMAL** | Mode + counts | Routing | 100-500 tokens |
| **METADATA** | Stats only | Synthesis | 50-200 tokens |

For tier selection guidance, see `references/context-tiers.md`.

## Quick Reference: Input Section Pattern

### Before (Anti-pattern)
```yaml
## Input
You receive:
- snapshot: Full context snapshot
- all_findings: Complete list
- full_config: Everything
```

### After (SOTA Pattern)
```yaml
## Input
You receive (SELECTIVE context):
- analysis_summary: Key findings only
- relevant_files: Files for this focus area
- mode: Analysis depth setting

**NOT provided** (context isolation):
- Full plugin contents
- Unrelated analysis results
- Other agents' intermediate work
```

## Anti-Patterns to Avoid

| Anti-Pattern | Problem | Fix |
|--------------|---------|-----|
| Snapshot Broadcasting | Same data to every agent | Tier by role |
| Defensive Inclusion | "Maybe they need this" | Document NOT PASSED |
| Grounding Everything | Validating low-priority | Severity batching |
| Large Embeddings | Full arrays when counts suffice | Reference pattern |
| Repeated Context | Same data multiple times in chain | Pass once, reference later |

## Handoff Protocol

Standard handoff between agents:

```yaml
handoff:
  from_agent: coordinator
  to_agent: analyzer
  context_level: SELECTIVE

  payload:
    mode: deep
    analysis_summary:
      claim_count: 15
      high_risk_count: 4
    relevant_files:
      - file: "[path]"
        content: "[content]"

  not_passed:
    - full_snapshot
    - unrelated_files
    - other_agents_data

  expected_output:
    format: yaml
    schema: AnalysisOutput
```

For complete handoff patterns, see `references/handoff-protocols.md`.

## Severity-Based Batching

Reduce validation operations by priority:

```yaml
batching:
  HIGH:     [all_validators]    # 4 agents
  MEDIUM:   [checker, estimator] # 2 agents
  LOW:      [checker]            # 1 agent
  INFO:     []                   # Skip

# Result: 60-70% fewer validation operations
```

## Metrics to Track

| Metric | Target | Calculation |
|--------|--------|-------------|
| Tier Compliance | 100% | Agents with tier / Total agents |
| Redundancy Ratio | < 0.1 | Duplicate data / Total data |
| Context per Agent | < 2K | Avg tokens per agent |
| NOT PASSED Coverage | 100% | Agents with exclusions / Total |

## Additional Resources

- `references/four-laws.md` - Detailed law explanations with examples
- `references/context-tiers.md` - Tier definitions and selection guide
- `references/handoff-protocols.md` - YAML schema patterns
- `references/examples.md` - Production examples from red-agent

Files: 5

Size: 27.5 KB

Complexity: 46/100

Category: AI Agents

Source: https://github.com/abossenbroek/abossenbroek-claude-plugins/tree/main/context-engineering/skills/context-engineering

Related in AI Agents

skill-development

Included

Comprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.

AI Agentsscripts

reprompter

Included

Transform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.

AI Agentsscripts

adaptive-compaction

Included

Adaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.

AI Agentsscripts

agent-skill-creator

Included

Create cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.

AI Agentsscripts

llm-wiki

Included

Use when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.

AI Agentsscripts

skill-master

Included

Agent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.

AI Agentsscripts