This roadmap reflects the current direction of PUDO. Priorities may shift based on community feedback. Items marked ✅ are shipped. Items marked 🔄 are in progress. Unmarked items are planned.
Focus: stability, completeness, and team adoption.
| Item | Status |
|---|---|
Prompt linter (pudo lint) |
✅ Shipped |
JSON output for check and doctor (--json flag) |
✅ Shipped |
| Copilot cloud agent setup template | ✅ Shipped |
| Run trace and metrics schemas | ✅ Shipped |
Team adoption guide (docs/team-adoption-guide.md) |
✅ Shipped |
| Publishable MCP server package with client config examples | ✅ Shipped |
| Claude Fable 5 support (1M context window, long-running tasks) | ✅ Shipped |
| Additional measured case studies | 🔄 In progress |
docs/getting-started.md first-steps guide |
🔄 In progress |
docs/architecture.md with system diagrams |
🔄 In progress |
| Expanded FAQ (team handoff, enterprise patterns) | Planned |
SUPPORT.md community channels |
✅ Shipped |
Focus: agent interoperability and broader ecosystem reach.
| Item | Notes |
|---|---|
| MCP remote transport and authentication guidance | Enables hosted / cloud-based agent use |
| Stack recipes with command detection | Auto-detect Next.js, FastAPI, Go, etc. |
| OpenAI Agents, LangGraph, and other framework adapters | Expand beyond coding-agent toolchains |
| Automated release workflow (version bump → publish) | Reduce manual release risk |
| Multi-version Node.js CI matrix (20.x, 22.x) | Catch compatibility regressions earlier |
| npm public registry publish for MCP server | Improve discoverability (npm install @pudo/mcp-server) |
Dependency audit CI step (npm audit) |
Supply chain safety in every PR |
Focus: enterprise readiness and ecosystem maturity.
| Item | Notes |
|---|---|
| Enterprise adoption guide | Compliance, governance, regulated environments |
| Metrics framework and dashboard | Visualize token waste, PUDO score trends over time |
| Agent skills and subagent workflows | Deep integration with emerging agent frameworks |
| GitHub App / bot for PUDO gate enforcement | Automated gate checks as a required status check |
| Community prompt registry | Searchable, rated, versioned community prompts |
| Benchmark dataset (10+ measured cases) | Validate directional targets with real evidence |
| Governance model (maintainer council) | Reduce single-maintainer risk |
Ideas under consideration — not committed, not scheduled:
A full history of shipped features is in CHANGELOG.md.
Key milestones:
pudo init), stack templates, benchmark kit, session handoffpudo check/score/doctor, MCP server alpha, agent tool security, CODEOWNERS, GitHub Actions