Skip to content
SkillModel-led3,206 tokens in use

context-engineering

Control what an AI coding agent sees with rules files, a context hierarchy, packing strategies and budget management.

119,377 stars on the repository MIT Licenseaddyosmani/agent-skills · skills/context-engineering
npx skills add addyosmani/agent-skills --skill context-engineering
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing context-engineering costs

44tokens idle
3,206tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

shelldetectedcredentialsdetected

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter12/15
  • Structure19/25
  • Examples15/15
  • Output15/15
  • Size15/15
  • Resources10/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent
  • Reads credentials−15

    does this on the machine, and the frontmatter does not say so

    SKILL.md, line 65

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community76
  • Maturity32
  • Momentum40

Context Engineering teaches an agent and its user to curate the information in a session. It ranks context from rules files down to conversation history and says what to load at each level.

The skill covers restartable session boundaries, three packing strategies, context budget management, MCP integrations and patterns for surfacing conflicts and missing requirements.

What You Get

  • A five-level context hierarchy
  • A rules file template and the names of the equivalent files for other agents
  • Packing strategies: brain dump, selective include and hierarchical summary
  • Restartable session rules, anti-pattern table and a verification checklist

Common Use Cases

  • Starting a coding session with the right context
  • Recovering when output quality drops
  • Writing a rules file for a project
  • Handing work to a fresh session

What Makes It Unique

The skill adds session boundary rules that name what to persist before a restart, a recency rule that places the active task material last, and explicit trust levels for loaded files. It also gives a format for raising spec conflicts to the user in place of silently choosing.

Under The Hood

The skill is a single SKILL.md of instructions and templates. It runs no scripts. It points the agent to read rules files, specs, source files and git status, and it lists MCP servers such as Context7, Chrome DevTools, PostgreSQL, Filesystem and GitHub that supply extra context.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers