The Skills Library is the largest and most frequently used component of ECC. Where rules encode constraints that are always active and agents handle delegated tasks, skills are on-demand knowledge modules — reusable workflow bundles your harness loads when context calls for them. Skills cover every major language stack, security discipline, agentic pattern, ML engineering workflow, and a growing set of domain verticals. Install the ones your project needs; the rest stay out of your context window.Documentation Index
Fetch the complete documentation index at: https://mintlify.com/affaan-m/ECC/llms.txt
Use this file to discover all available pages before exploring further.
What Is a Skill?
A skill is a directory (or a single.md file) installed to ~/.claude/skills/ — or the equivalent path for your harness target. Each skill contains at minimum a SKILL.md with a YAML frontmatter block, a purpose statement, activation triggers, workflow steps, and copy-pasteable examples.
Skills activate when:
- The user’s task matches the skill’s declared domain
- The harness detects relevant file types or imports
- A command or agent explicitly references the skill
- You name the skill in your prompt
commands/ layer is best understood as legacy slash-entry compatibility during migration; the canonical workflow definition lives in the underlying skill.How Skills Are Organized
ECC groups skills into modules, and modules are addressed by component IDs. A component ID likelang:typescript or capability:security maps to one or more modules; each module lists the exact skill directories that get installed.
The module families are:
| Family | Component prefix | Examples |
|---|---|---|
| Language | lang: | lang:typescript, lang:python, lang:rust |
| Framework | framework: | framework:react, framework:django, framework:rails |
| Capability | capability: | capability:security, capability:agentic, capability:machine-learning |
| Skill (direct) | skill: | skill:tdd-workflow, skill:deep-research, skill:eval-harness |
Skill Module Categories
ECC’s install manifest ships over 20 skill modules plus language and framework packs. The major categories:workflow-quality — Quality, TDD, and Learning
workflow-quality — Quality, TDD, and Learning
tdd-workflow, eval-harness, verification-loop, strategic-compact, continuous-learning, continuous-learning-v2, windows-desktop-e2e, e2e-testing, ai-regression-testing, agent-introspection-debugging, production-audit, skill-scout, skill-stocktake, and more.framework-language — Languages and Frameworks
framework-language — Languages and Frameworks
security — Security Review and Compliance
security — Security Review and Compliance
agentic-patterns — Agentic Engineering
agentic-patterns — Agentic Engineering
machine-learning — ML Engineering Workflows
machine-learning — ML Engineering Workflows
mle-workflow skill.research-apis — Research and Investigation
research-apis — Research and Investigation
deep-research, Exa search integration, research operations, and scientific database access for PubMed and USPTO.database — Persistence Skills
database — Persistence Skills
business-content — Business and Content
business-content — Business and Content
devops-infra — DevOps and Infrastructure
devops-infra — DevOps and Infrastructure
swift-apple — Swift and Apple Platforms
swift-apple — Swift and Apple Platforms
optimization-workflows — Performance and Optimization
optimization-workflows — Performance and Optimization
operator-workflows — Operator Integrations
operator-workflows — Operator Integrations
supply-chain-domain — Supply Chain
supply-chain-domain — Supply Chain
media-generation — Media and Video
media-generation — Media and Video
Finding and Installing Skills
Browse the catalog
AI-powered recommendation
Install by component ID
Install by profile
~/.claude/skills/ (or the equivalent path for your harness). They become available immediately in your next session.
Using Skills in Your Harness
Once installed, skills activate in two ways: Automatic activation — the harness loads a skill when your task context matches the skill’s declareddescription and When to Activate section. For example, opening a .py file and asking to add tests can trigger python-testing and tdd-workflow automatically.
Explicit reference — name the skill or its domain in your prompt: