agentsclimarketplace

Context engineering

Skill Byunk/minimal-claude-code/context-engineering/skills/context-engineering

A minimal Claude Code plugin with just the essentials.

Install
npx -y skills add Byunk/minimal-claude-code --skill context-engineering

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 9 stars9 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Guidance for designing, auditing, and improving context supplied to LLM agents, prompts, tools, RAG/MCP servers, and multi-agent systems. Use when the task involves context selection, prompt or agent architecture, tool descriptions, retrieval context, memory, compaction, autonomy, or evaluation of LLM behavior.

SKILL.md

4.0 KB, as published. Nobody here has run it

Context Engineering

A practical guide for choosing, structuring, and evaluating the context supplied to LLM systems. It covers prompts, tool descriptions, retrieval, memory, compaction, autonomy, and agent architecture.

References

#PrincipleFileSummary
1Quality Over Quantity01-quality-over-quantity.mdInclude only relevant context. Precision beats exhaustiveness.
2Start Minimal, Layer Progressively02-start-minimal.mdBegin with essentials, expand based on performance gaps.
3Hierarchical Context Organization03-hierarchical-organization.mdStructure context from general to specific. Use XML tags for disambiguation.
4Multi-Dimensional Context04-multi-dimensional-context.mdEffective context spans task, domain, history, constraints, motivation, and format.
5Simple Composable Patterns05-simple-composable-patterns.mdPrefer prompt chaining, routing, parallelization over complex frameworks.
6Invest in Tool Design (ACI)06-invest-in-tool-design.mdTool descriptions deserve as much effort as UIs. Calibrate action language for Claude 4.6.
7Match Architecture to Complexity07-match-architecture-to-complexity.mdChoose the right pattern: chaining, routing, parallelization, orchestrator, or agent.
8Transparency in Agent Planning08-transparency-in-planning.mdShow planning steps. Use structured state tracking for long-horizon tasks.
9Contextualize Retrieved Chunks09-contextualize-retrieved-chunks.mdPrepend explanatory context before embedding. Reduces retrieval failures by 35-67%.
10Dynamic Context Adaptation10-dynamic-context-adaptation.mdAdjust context based on task complexity and model confidence.
11Meta-Context11-meta-context.mdExplain why information is provided. Use grounding directives to reduce hallucinations.
12Ground Agents in Environment Feedback12-ground-in-environment.mdUse tool results and human checkpoints, not just model reasoning.
13Test and Measure Systematically13-test-and-measure.mdA/B test context variations. Measure accuracy, latency, reasoning quality.
14Retrieve Generously, Filter Late14-retrieve-generously.mdTop-20 chunks outperform top-5. Don't over-filter prematurely.
15Avoid Common Anti-Patterns15-anti-patterns.mdDon't dump everything, over-abstract, overengineer, or inherit stale aggressive prompts.
16Guide Model Thinking16-guide-model-thinking.mdControl reasoning depth with adaptive thinking. Prevent overthinking on simple tasks.
17Multi-Window State Management17-multi-window-state.mdExternalize state for tasks spanning multiple context windows.
18Calibrate Autonomy and Safety18-calibrate-autonomy.mdTune action thresholds. Claude 4.6 defaults to high autonomy; set guardrails.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.