Developer + agent intelligence layer. Combines the DevPlane developer schema (apps, packages, services, environments, deployments, recipes, decisions, incidents) with the OpenClaw delegation primitives (project, task, skill, provider). Agents are pod-level identities — not tracked as workspace entities.
Record types
The structured data this adds to your pod.
App
A software application in your development stack
- App Name
- Description
- Status
- Repository URL
- Deploy URL
- Tech Stack
- Dev Port
- AI Channel ID
- +4
Feature
A feature or work item with a persistent AI session
- Status
- Priority
- Linked App
- Linked Packages
- Linked Environment
- Estimated Days
- AI Channel ID
- Agent Status
- +11
Story
A user-facing story that orchestrates features and drives the AI pipeline
- Status
- Priority
- Linked App ID
- Linked Feature IDs (comma-separated)
- Linked Package IDs (comma-separated)
- AI Channel ID
- Agent Status
- Context Doc ID
- +6
Service
A backend service or microservice
- Status
- Type
- Host
- Port
- Health Check URL
- Linked App
- Linked Server
- Repo Local Path
Package
An npm/pnpm package — tracked with health, graph edges, and publish history
- Package Name
- Version
- Published
- Registry
- Depends On
- Consumed By Apps
- Health Score
- Health Grade
- +5
Server
A server or deployment target — SSH-accessible, provider-agnostic
- Environment
- Host / IP
- App URL
- Active
- SSH User
- Linked App
- Role
- Provider
Deployment Target
A deployment provider target - Coolify, Vercel, or custom
- Provider Type
- API URL
- Auth Token Ref
- Environment
- Status
- Linked App Slug
Deployment
A deployment event or release
- Status
- Version
- Commit SHA
- Linked App
- Environment
- Deployed By
- Deployed At
- Logs URL
- +1
Recipe
An ordered list of shell commands to run on a remote server
- Recipe Name
- Description
- Template
- On Failure
- Recipe Steps
- Linked Environment
- Linked App
- Rollback Recipe
Recipe Run
A single execution of a recipe with per-step status tracking
- Recipe
- Status
- Run Steps
- Started At
- Finished At
- Duration (ms)
- Triggered By
- Linked App
Prompt Snippet
A reusable AI prompt template
- Title
- Category
- Description
- Prompt Template
- Executable
- Linked App
Decision Record
A technical decision with context, rationale, and alternatives
- Context
- Decision
- Alternatives Considered
- Linked App
- Linked Packages
- Made By
- Made At
Incident
A production incident with severity, status, and root cause tracking
- Severity
- Status
- Affected Apps
- Root Cause
- Resolved At
- Description
- Linked Environment
- Recipe Run
Cron Job
A scheduled task with cron expression and run history
- Cron Expression
- Status
- Description
- Linked App
- Last Run At
- Last Run Status
- Next Run At
Package Publish
A recorded npm publish event with version, changelog, and consumer impact
- Package
- From Version
- To Version
- Bump Type
- Changelog
- Status
- Consumers Notified
- Triggered By
- +1
Agent Run
A single pipeline execution of an autonomous agent for a feature
- Feature ID
- Pipeline Phase
- Run Status
- Agent
- Started At
- Ended At
- Duration (ms)
- Output Summary
- +2
Best Practice
A codified engineering best practice for a tech stack or workflow
- Category
- Tech Stack
- Title
- Content
- Source
- Applies To Apps
- Enforced by CI
Codebase Map
A structural map of a module or directory within an app
- App
- Path
- Module Name
- Description
- Owner
- Tech Tags
- Key Files
- Conventions
- +1
Project
A bounded body of work delegated to one or more agents
- Status
- Priority
- Human Owner
- Start Date
- Deadline
- Health
- Type
- Goal
- +1
Task
An atomic work item assigned to a coder agent or background runner
- Status
- Priority
- Project
- Type
- Due Date
- Effort Estimate (hrs)
- Token Estimate (K)
- Actual Tokens Used (K)
- +2
Skill
An installed agent capability — a ClawHub skill, custom script, or API integration
- Trust Level
- Category
- Source URL
- Version
- Install Date
- Last Audit Date
- VirusTotal Status
- Maintainer
- +2
Provider
An LLM or API provider — tracks cost, key health, and agent usage
- Type
- Key Health
- Model / Endpoint
- Cost per 1K Input Tokens ($)
- Cost per 1K Output Tokens ($)
- Daily Budget Ceiling ($)
- Actual Spend This Week ($)
- Base URL / Endpoint
- +1
System Component
A building block of your architecture — frontend, backend, database, service, or external dependency
- Type
- Status
- Tech Stack
- Repository URL
- Deploy / Dashboard URL
- Linked App ID
- Linked Package IDs
- Health Score
- +2
Workflow
A user flow or business process — a sequence of steps through system components
- Status
- Category
- Trigger
- Expected Outcome
- Linked App ID
- Linked Component IDs
- Mermaid Diagram
- Description
Capability
A system capability — what the system can do (rendering, data, auth, AI, infra, UI, workflow)
- Type
- Maturity
- Description
- Owner
- Code Path
- Linked App IDs
- Linked Package IDs
- Linked Component IDs
Views
- AppsTable
- Task BoardBoard
- DeploymentsCalendar
- PackagesTable
- IncidentsTable
- ProjectsBoard
- Architecture GraphFlow
- WorkflowsFlow
- Capability MapFlow
Playbooks
Repeatable runs your agent can start for this project.
- Evaluate FeatureAssess a feature — its current state, whether it's worth building, and how it fits the rest of the system.
Relationships
- Devplane workflow Devplane system componentTraverses
- Devplane workflow Devplane capabilityExercises
- Devplane capability Devplane system componentProvided by
- Devplane system component Devplane system componentDepends on
First conversation
After install, your agent sets the project up with you. It starts here:
Capture the real engineering surface — the apps that are shipped, where their code lives, what runs behind them, where they deploy, and what's in flight — so the workspace mirrors what the team actually builds and runs, not a generic templ…
- “What have you actually shipped, and what's still just planned? Give me the apps and their repos.”
- “For each app — what runs behind it (services, workers), and where does it deploy?”
- “What's on your plate this week, and which of it should be delegated to agents?”