AOSP search target extraction and RCA hypothesis synthesis specialist
Strategic Architecture & Debugging Advisor (Opus, READ-ONLY)
Work plan and code review expert — thorough, structured, multi-perspective (Opus)
AOSP code investigation specialist using remote AOSP MCP search
Android log collection specialist — downloads/unpacks/classifies logs from JIRA attachments or local directories, producing extracted/ and file-classification.json for downstream parsing
General AOSP source/module/function/build-artifact technical report — analyze a feature, module, function, generated file, build artifact, or subsystem via AOSP source search and produce a structured technical report.
AOSP multi-repository execution engine that parses cross-repository change plans produced by aosp-plan and runs them in parallel
Export a FOSD-style meta-slice package for vendor-modified or vendor-added AOSP features
Import a six-layer AOSP vendor feature meta-slice package into another AOSP project and produce a plan-ready mapping guide
AOSP investigation-driven planning with consensus review
Modifies files
Hook triggers on file write and edit operations
Requires secrets
Needs API keys or credentials to function
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.
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
@zeonic/zaku — 量産型開発支援兵器
Mass-production dev weapon for Claude Code and OpenAI Codex — AOSP recon, JIRA elimination, Git ammo supply.
Granada is Zeonic's orbital fortress—where Zaku mobile suits are mass-produced before they sortie into combat. This project ships the same Android platform workflows to Claude Code and Codex: AOSP source search, JIRA-driven root-cause analysis, and Git workflow automation.
Node.js 20 or newer is required for the bundled MCP bridge and lifecycle hooks.
npx skills add @zeonic/zaku
This installs the published skill workflows. For the complete local plugin surface, including agents, hooks, and MCP servers, clone the repository and install it as a Claude Code plugin:
git clone https://github.com/INNERJOINT/Granada.git
cd Granada && claude plugin install .
Claude Code discovers the repository's canonical skills/, agents/, hooks/hooks.json, and .mcp.json surfaces directly.
Install Granada as a repository marketplace, then install the lightweight Zaku plugin from that marketplace:
git clone https://github.com/INNERJOINT/Granada.git
cd Granada
codex plugin marketplace add "$PWD"
codex plugin add zaku@zeonic-local
Confirm that Codex sees the plugin and its optional MCP servers:
codex plugin list --json
codex mcp list
Start a new Codex thread after installing or updating the plugin so its generated skills, hooks, and MCP configuration are rediscovered. The marketplace intentionally points at plugins/zaku instead of the repository root, keeping local Git state, dependencies, and host-specific configuration out of the Codex plugin cache.
The workflow name is shared across hosts; only the invocation prefix changes:
| Host | Example |
|---|---|
| Claude Code | /zaku:aosp-plan "trace the Android boot flow" |
| Codex | $zaku:aosp-plan "trace the Android boot flow" |
Generated Codex workflows translate Claude-specific delegation and tool notation to Codex-native skills, collaboration tools, and MCP names. Codex keeps zaku: as the plugin skill namespace, but uses $ instead of Claude's / prefix.
| Skill | Description |
|---|---|
aosp-analyze | General AOSP module/function/feature technical report |
aosp-rca | Android crash/issue root-cause analysis via AOSP source search |
aosp-plan | Investigation-driven AOSP planning with parallel code search |
aosp-autopilot | Multi-repo automatic execution engine |
aosp-feature-export | Export AOSP feature element documentation |
aosp-feature-import | Import AOSP features across projects |
aosp-project | List SourcePilot projects and select the active AOSP project |
jira-analyze | JIRA-driven Android bug RCA with parallel hypothesis investigation |
jira-aftersales | Convert RCA reports to customer-facing scripts |
gitlab-info | Extract repository, commit, changed files, and diffs from GitLab URLs |
git-commit | Generate commit messages following repo conventions |
Proxies requests to a remote AOSP MCP server for code search across multiple Android projects. Set SOURCEPILOT_URL and, when required, SOURCEPILOT_KEY in the environment that starts the host.
The Codex bundle uses a dependency-free stdio proxy. It mirrors the remote SourcePilot tool definitions dynamically; if the server is not configured or discovery fails, it still exposes a stable fallback schema and returns a clear configuration or connection error when a tool is called.
MCP tool namespaces differ by host:
| Server | Claude Code | Codex | Environment |
|---|---|---|---|
| SourcePilot | mcp__plugin_zaku_sourcepilot__* | mcp__sourcepilot__* | SOURCEPILOT_URL, SOURCEPILOT_KEY |
| Atlassian | mcp__plugin_zaku_atlassian__* | mcp__atlassian__* | JIRA_URL, JIRA_PERSONAL_TOKEN, CONFLUENCE_URL, CONFLUENCE_PERSONAL_TOKEN |
| GitLab | mcp__plugin_zaku_gitlab__* | mcp__gitlab__* | GITLAB_PERSONAL_ACCESS_TOKEN, GITLAB_API_URL |
The optional Atlassian and GitLab stdio servers also require uvx and npx, respectively, on PATH.
npx claudepluginhub innerjoint/granada --plugin zakuMulti-agent orchestration system for Claude Code
Evidence-gated AI coding workflow: scan → analyze → plan → TDD → execute → fix → verify → review, powered by Codebase Memory MCP >= 0.9.0 with optional Serena LSP intelligence. Includes blast-radius planning, test/cycle gates, independent review, and Windows Git Bash hook auto-resolution.
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.
Consult multiple AI coding agents (Gemini, OpenAI, Grok, Perplexity, plus codex, antigravity, and grok CLIs when installed) to get diverse perspectives on coding problems
Production-grade engineering skills for AI coding agents — covering the full software development lifecycle from spec to ship.
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.