Spec-driven coding workflow plugin. Wires Jira → OpenSpec → GSD/Superpowers → code → Jira, with per-change token tracking, model fallback routing, progress aggregation, session/process resume, and named checkpoints. Seven auto-triggering skills, eleven commands, a statusline hook with progress and routing warnings, plus a unified .mina/ state layout.
Save or restore a named workflow state snapshot
List Jira issues from my queue and pick one to start work on
Convert a Jira issue into an OpenSpec change proposal
Update the linked Jira issue after implementation completes
Show current model, recommended model for active context, and reason
Use this skill whenever the user asks about Jira issues, picking a ticket, listing their assigned work, transitioning status, or commenting on a ticket — AND the Atlassian MCP server is unavailable, returning auth errors, or has been disabled. Uses the official Atlassian CLI (acli) as a reliable fallback. Also use when the user explicitly asks for "acli" or says MCP is broken.
Use this skill whenever errors occur related to rate limits (429), overload (529), 5-hour window exhaustion, "rate limited" or "waiting for capacity" messages, or whenever context exceeds 80%, cost on a single change exceeds $5, or the user explicitly asks about switching models or fallback. Defines the project's model routing tiers and the decision rules for when to switch.
Use this skill BEFORE any planning, brainstorming, write-plan, or execute-plan activity in a project that has an openspec/ directory. Checks for existing OpenSpec change proposals that match the user's request and treats them as authoritative input, preventing duplicate planning work and spec drift. Activates automatically when the user asks to plan, build, implement, or start work on a feature in a spec-driven project.
Use this skill whenever the user asks to resume work, pick up where they left off, continue a previous session, restart an interrupted task, or asks about background processes (dev servers, watch builds, long-running tests). Also activates when the user mentions session IDs, `claude --resume`, `--continue`, or "yesterday's work". Defines how to map prior sessions/changes/processes to a clear resume path.
Use this skill whenever the user asks about progress, status, what's done, what's left, where things stand, "what was I doing", "how far along", or asks for a summary of current work. Also activates when starting a new session in a project with .mina/ state, to surface relevant context. Defines how the plugin aggregates progress across OpenSpec tasks, GSD plans, git commits, and cost data into a unified view.
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.
A Claude Code plugin marketplace and opencode-compatible skill bundle for spec-driven development. Wires together Jira → OpenSpec → GSD/Superpowers → code → Jira into a single pipeline.
/plugin marketplace add astronaut1712/astronaut-ai
/plugin install mina@mina
Restart Claude Code. Verify with /help — you should see /mina:jira-pick and the other three commands.
git clone https://github.com/astronaut1712/astronaut-ai.git
cd astronaut-ai
chmod +x install.sh
./install.sh opencode # project scope
./install.sh opencode --user # user scope (~/.config/opencode/)
If you want to edit the skills/commands directly in your project (e.g. customize for your team's stack), skip the marketplace and use the install script:
./install.sh claude-code # project scope, into .claude/
./install.sh claude-code --user # user scope, into ~/.claude/
./install.sh both # both Claude Code and opencode
The trade-off: manual install gives flat command names (/jira-pick), marketplace gives namespaced (/mina:jira-pick).
astronaut-ai/
├── .claude-plugin/
│ └── marketplace.json ← marketplace catalog
├── plugins/
│ └── mina/ ← the plugin
│ ├── .claude-plugin/
│ │ └── plugin.json
│ ├── skills/
│ │ ├── spec-driven-workflow/
│ │ ├── openspec-aware/
│ │ ├── jira-via-acli/
│ │ ├── token-tracking/
│ │ ├── model-fallback/
│ │ ├── progress-tracking/
│ │ └── process-resume/
│ ├── commands/
│ │ ├── jira-pick.md
│ │ ├── jira-to-spec.md
│ │ ├── spec-to-plan.md
│ │ ├── jira-update.md
│ │ ├── token-report.md
│ │ ├── model-route.md
│ │ ├── model-switch.md
│ │ ├── status.md
│ │ ├── resume.md
│ │ ├── processes.md
│ │ └── checkpoint.md
│ ├── hooks/
│ │ └── statusline.sh ← progress + cost + routing warnings
│ └── README.md
├── templates/
│ ├── CLAUDE.md.snippet
│ ├── mcp.json.example ← Atlassian MCP config
│ ├── settings.json.example ← Claude Code statusline wiring
│ ├── model-routing.json.example ← model tiers + fallback chains
│ ├── opencode.json.example ← opencode multi-provider config
│ └── proposal-template.md ← OpenSpec proposal with Jira frontmatter
├── install.sh ← opencode + manual installer
├── setup-and-push.sh ← personalize + git init
├── README.md ← you are here
├── LICENSE
├── CHANGELOG.md
└── .gitignore
The project also expects .mina/ to be created at runtime in your working project:
<your-project>/.mina/
├── state.json ← active change/phase/plan/sessions/processes/checkpoints/history
├── tokens/<change>.jsonl ← per-change cost logs
└── checkpoints/<name>.json ← named state snapshots
Required regardless of install path:
Add project rules — append templates/CLAUDE.md.snippet to your project's CLAUDE.md (Claude Code) or AGENTS.md (opencode)
Set up Atlassian credentials
# Create API token at https://id.atlassian.com/manage-profile/security/api-tokens
export ATLASSIAN_AUTH=$(echo -n "[email protected]:your-api-token" | base64)
# Add to ~/.zshrc or ~/.bashrc to persist
Then copy templates/mcp.json.example to .mcp.json at your project root.
Install the runtime tools (the plugin is glue, not the engine):
npm install -g @fission-ai/openspec # required
npx get-shit-done-cc@latest # if using GSD execution
# Superpowers via /plugin in Claude Code
brew install --cask acli # optional Jira fallback
Initialize OpenSpec in your project:
openspec init
(Optional) Enable cost tracking statusline — see the next section.
The plugin logs token usage per OpenSpec change so you can attribute spend to specific work, not just sessions. This is opt-in and requires a one-time statusline setup.
Copy templates/settings.json.example to ~/.claude/settings.json (user-wide) or <project>/.claude/settings.json (per-project). The example wires:
{
"statusLine": {
"type": "command",
"command": "${CLAUDE_PLUGIN_ROOT}/hooks/statusline.sh",
"padding": 0
}
}
Restart Claude Code. You'll see a statusline like:
🤖 opus | 💰 $0.23 | 🧠 45% | 📝 ENG-1234 feat-add-dashboard-ssr
npx claudepluginhub astronaut1712/astronaut-ai --plugin minaComprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
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.
Harness-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
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.