# ECC

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Repository: https://github.com/affaan-m/ECC
Canonical: https://ross.abutalabs.com/products/ecc
Homepage: https://ecc.tools
Language: JavaScript
License: MIT
License Family: permissive
Topics: ai-agents, anthropic, claude, claude-code, developer-tools, llm, mcp, productivity
Last push: 2026-08-25T21:27:27+00:00
Link (homepage): https://ecc.tools
Link (site_page): https://ecc.tools/pricing
Link (site_page): https://ecc.tools/skills
Link (site_page): https://ecc.tools/changelog

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 95, longevity 16
- inputs: {"age_days": 228, "days_push": 8, "days_rel": 37, "gap_med": 6, "n_releases_24m": 14}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 243524, forks 36832 (observed 2026-08-28T04:12:28.661106+00:00)

## What it is
ECC is an open-source 'agent harness operating system' that provides 261 skills, 64 agents, and 84 commands to optimize AI coding agents like Claude Code, Codex, Cursor, and OpenCode. It bundles workflow skills, continuous-learning instincts, memory, and AgentShield security scanning, with an optional commercial GitHub App layer for repo analysis and team rollout.

## Use cases
- optimize my claude code agent performance
- add reusable skills and commands to my coding agent
- scan agent sessions for security risks
- turn repo history into reusable agent defaults
- set up tdd and verification workflows for ai coding agents
- make my coding agent learn from past sessions
- manage mcp servers and agent context across tools

## When to choose
- you use Claude Code, Codex, Cursor, or OpenCode daily and want structured skills, memory, and security guardrails
- you want a curated catalog of agent skills instead of writing your own prompts and workflows
- you need session-level security scanning and learned instincts with confidence scoring
- you want a GitHub App that converts repo conventions into reusable agent defaults

## When to avoid
- you don't use AI coding agents and just need a plain linter or CI tool
- you need a lightweight setup and don't want a large skill/agent catalog in your repos
- you require private-repo GitHub App analysis without a paid plan
- you prefer fully self-contained tooling with no commercial upsell layer

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, mcp, security, developer-tools, prompt-engineering, workflow-automation, code-review
- domain: developer-tools, security
- platform: cli, cross-platform, cloud
- tags: claude-code, codex, cursor, opencode, agent-harness, skills, instincts, agentshield, github-app, context-management, tdd-workflow, eval-harness, ai-agents, automation, nodejs, docker

## Member repositories
- affaan-m/ECC (main) score 81
- affaan-m/agentshield (plugin) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:28.661106+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:10:20.536435+00:00, confidence not recorded.
  - readme: https://github.com/affaan-m/ECC (fetched 2026-08-28T04:12:28.661106+00:00, sha 98bdb870e8ce)
  - homepage: https://ecc.tools (fetched 2026-08-28T17:21:29.069465+00:00, sha 0774380ef937)
  - site_page: https://ecc.tools/pricing (fetched 2026-08-28T17:21:29.078551+00:00, sha 276e8282fae1)
  - site_page: https://ecc.tools/skills (fetched 2026-08-28T17:21:29.080348+00:00, sha d9574a435d3d)
  - site_page: https://ecc.tools/changelog (fetched 2026-08-28T17:21:29.082324+00:00, sha ed1e85d0fd8e)
- Data as of 2026-08-30T08:39:29.467469+00:00.