By MrShininnnnn
All the context, in one place for Claude Code, Codex, Gemini, and AI coding tools.
Use when the user wants to start tracking a local custom instruction file (like .agy/ANTIGRAVITY.md or .github/copilot-instructions.md) in SuperCtx. Requires a path argument.
Use when setting up SuperCtx in a project for the first time, or when the user asks to create the .ctx hub, connect existing instruction files, or "initialize superctx". Also use when no .ctx/ directory exists yet but the user wants cross-tool context management.
Use when the user wants a SuperCtx status check, asks about hub-and-shim health, or needs to verify that the hub exists and registered shims/backups are healthy.
Use when the user wants to repair or regenerate SuperCtx shims, or when /superctx:status reports missing or broken registered shim files.
Use when entering a repository with a .ctx/ hub, or when the user asks about project context, agent instructions, tool configuration, health, candidates, broken shims, repair, or cross-tool context.
Modifies files
Hook triggers on file write and edit operations
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
All the context, in one place.
SuperCtx keeps AI coding assistants aligned on one shared project context. It creates a project-local .ctx/ hub from files like CLAUDE.md, AGENTS.md, and GEMINI.md, then keeps tool-specific instruction files as thin generated shims pointing back to that hub.
Many coding tools use their own project instruction files:
CLAUDE.md
AGENTS.md
GEMINI.md
.github/copilot-instructions.md
When you use more than one tool, those files can drift apart. SuperCtx gives the project one context layer and exposes it consistently across Claude Code, Codex, Gemini, Antigravity, Copilot, Cursor, and other AI coding tools.
Use SuperCtx when a repository has more than one AI coding tool and you want those tools to share the same project assumptions.
SuperCtx helps you:
SuperCtx uses a project-local context folder:
.ctx/
The .ctx/ folder is the centralized hub for project context. SuperCtx makes .ctx/SUPERCTX.md the canonical hub, and generates thin shims pointing back to it:
┌─> CLAUDE.md (shim)
.ctx/SUPERCTX.md ┼─> AGENTS.md (shim)
└─> GEMINI.md (shim)
SUPERCTX.md is a version-controlled Markdown file and the canonical shared context hub. You can inspect or edit it directly if needed, but the normal workflow is to ask the agent to update project context. Original tool instruction files are backed up and converted to thin referential shims.
A built-in write guard also intercepts direct edits/writes to protected instruction files, guiding the agent to update the shared context hub instead.
SuperCtx includes a built-in agent reference skill, /using-superctx. This is an agent reference / advanced skill. Normal users do not need to run this command; the agent uses it automatically to orient itself on the project's context boundaries and active files.
When SuperCtx is active, the agent uses .ctx/SUPERCTX.md as shared project context through generated shims. You normally ask the agent to update project context naturally instead of managing Markdown or TOML by hand.
When SuperCtx is actively helping, the agent may use a short cue such as:
Using SuperCtx to check context health.
The cue should stay task-specific and brief. SuperCtx should not explain hub, shim, or backup internals unless you ask.
SuperCtx scans the project directory during setup, status checks, session start, and protected instruction-file writes:
CLAUDE.md, AGENTS.md, GEMINI.md, and hidden standard paths like .claude/CLAUDE.md, .codex/AGENTS.md, and .github/copilot-instructions.md. These are automatically added to .ctx/manifest.toml..agents/rules/ and agents/skills/ are surfaced as candidates for future connection work..agent/rules/ or .agents/skills/ are reported..agy/ or .agy/ANTIGRAVITY.md are surfaced and labeled as unverified local candidates (not official support).SuperCtx is early software. Today, install it directly from GitHub for Claude Code or run the Python CLI from a source checkout.
From inside Claude Code, add this repository as a plugin marketplace and install the superctx plugin:
/plugin marketplace add MrShininnnnn/SuperCtx
/plugin install superctx@superctx
Claude Code copies marketplace plugins into Claude's local plugin cache. After installing or updating the plugin, restart Claude Code if the new skills do not appear immediately.
After installation, the primary command to check context health is:
/superctx:status
[!NOTE] SuperCtx is fully agent-guided. Use
/superctx:statusto check health. The agent will guide setup, add, and repair when needed. Normal users do not need to operate the other commands directly.
For advanced troubleshooting or developer verification, the following commands are available:
/superctx:init: Scaffold/migrate the context hub./superctx:sync: Repair generated shims./superctx:add <path>: Register a new local instruction file.npx claudepluginhub mrshininnnnn/superctx --plugin superctxHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review