mirror of https://github.com/affaan-m/everything-claude-code.git synced 2026-04-14 22:13:41 +08:00

Files

Affaan Mustafa dd1d505b9f feat: add lead-intelligence skill, autonomous-agent-harness, and Gemini CLI target

New skills:
- lead-intelligence: AI-native lead intelligence pipeline with 4 agents
  (signal-scorer, mutual-mapper, enrichment-agent, outreach-drafter).
  Replaces Apollo/Clay with agent-powered signal scoring, mutual ranking,
  warm path discovery, and personalized outreach drafting.
- autonomous-agent-harness: Replaces standalone agent frameworks (Hermes,
  AutoGPT) using Claude Code native crons, dispatch, memory, and computer
  use. Documents the full architecture for persistent autonomous operation.

New CLI target:
- Gemini CLI: .gemini/GEMINI.md config added, gemini target registered
  in install-modules.json platform-configs module.

Updated:
- marketplace.json: Fixed stale counts (was "14+ agents, 56+ skills"),
  now accurately reflects 30 agents, 138 skills, 60 commands.
- README.md and AGENTS.md: Synced skill counts to 138.
- install-modules.json: Added lead-intelligence to business-content
  module, autonomous-agent-harness to agentic-patterns module.

All catalog validations pass (30/60/138). Install-manifest tests pass (20/20).

2026-03-30 23:30:20 -04:00

1.8 KiB

Raw Blame History

ECC for Gemini CLI

This file provides guidance to Gemini CLI when working with code in projects that have ECC installed.

Overview

Everything Claude Code (ECC) is a production-ready AI coding plugin providing 28 agents, 127 skills, and 60 commands for software development workflows.

Core Workflows

Test-Driven Development

Write tests first (RED)
Run tests — they should fail
Write minimal implementation (GREEN)
Run tests — they should pass
Refactor (IMPROVE)
Verify 80%+ coverage

Code Review

Check for security vulnerabilities (OWASP Top 10)
Verify error handling at system boundaries
Ensure immutable data patterns
Validate input at API boundaries

Architecture

Many small files over few large files (200-400 lines, 800 max)
Organize by feature/domain, not by type
Repository pattern for data access
API response format: { success, data?, error?, meta? }

Coding Standards

Prefer immutability — create new objects, never mutate
Functions under 50 lines
No deep nesting (4 levels max)
Validate user input with schema validation (Zod, Joi, etc.)
No hardcoded secrets — use environment variables
Handle errors comprehensively with user-friendly messages

Security Checklist

Before any commit:

No hardcoded secrets
All user inputs validated
SQL injection prevention (parameterized queries)
XSS prevention
CSRF protection
Authentication/authorization verified

Commit Format

<type>: <description>

Types: feat, fix, refactor, docs, test, chore, perf, ci

Available Skills

Skills are located in the skills/ directory. Each contains a SKILL.md with detailed instructions.

Key skills: tdd-workflow, security-review, coding-standards, backend-patterns, frontend-patterns, lead-intelligence, autonomous-agent-harness, deep-research, exa-search, x-api

1.8 KiB Raw Blame History