Back to list
abossenbroek

context-engineering

by abossenbroek

0🍴 0📅 Jan 4, 2026

SKILL.md


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

LawPrincipleToken Impact
1. Selective ProjectionPass only fields each agent needs-30-50%
2. Tiered FidelityDefine explicit context tiers per role-40-60%
3. Reference vs EmbeddingUse references for large data-50-80%
4. Lazy LoadingLoad data on-demand, not upfront-30-50%

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

Context Tiers

TierDescriptionUse CaseTypical Size
FULLComplete dataInitial analysis5-20K tokens
SELECTIVERelevant subsetDomain workers1-5K tokens
FILTEREDCriteria-matchedValidators500-2K tokens
MINIMALMode + countsRouting100-500 tokens
METADATAStats onlySynthesis50-200 tokens

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

Quick Reference: Input Section Pattern

Before (Anti-pattern)

## Input
You receive:
- snapshot: Full context snapshot
- all_findings: Complete list
- full_config: Everything

After (SOTA Pattern)

## 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-PatternProblemFix
Snapshot BroadcastingSame data to every agentTier by role
Defensive Inclusion"Maybe they need this"Document NOT PASSED
Grounding EverythingValidating low-prioritySeverity batching
Large EmbeddingsFull arrays when counts sufficeReference pattern
Repeated ContextSame data multiple times in chainPass once, reference later

Handoff Protocol

Standard handoff between agents:

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:

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

MetricTargetCalculation
Tier Compliance100%Agents with tier / Total agents
Redundancy Ratio< 0.1Duplicate data / Total data
Context per Agent< 2KAvg tokens per agent
NOT PASSED Coverage100%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

Score

Total Score

50/100

Based on repository quality metrics

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

+10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新がある

0/10
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

0/5

Reviews

💬

Reviews coming soon