openwork-debug
$
npx mdskill add different-ai/openwork/openwork-debugDebug and inspect OpenWork configurations and audit trails using API and CLI tools
- Troubleshoot sidecar configurations, workspace health, and audit logs
- Uses OpenWork API endpoints and OpenCode CLI commands for diagnostics
- Analyzes server responses, config snapshots, and SQLite databases for insights
- Provides structured output via terminal and direct file inspection for developers
SKILL.md
.github/skills/openwork-debugView on GitHub ↗
--- name: openwork-debug description: Debug OpenWork sidecars, config, and audit trail --- ## Credential check Set these before running the HTTP checks: - `OPENWORK_SERVER_URL` - `OPENWORK_SERVER_TOKEN` - `OPENWORK_WORKSPACE_ID` (optional; use `/workspaces` to discover) ## Quick usage (read-only) ```bash curl -s "$OPENWORK_SERVER_URL/health" curl -s "$OPENWORK_SERVER_URL/capabilities" \ -H "Authorization: Bearer $OPENWORK_SERVER_TOKEN" curl -s "$OPENWORK_SERVER_URL/workspaces" \ -H "Authorization: Bearer $OPENWORK_SERVER_TOKEN" ``` ## Workspace config snapshot ```bash curl -s "$OPENWORK_SERVER_URL/workspace/$OPENWORK_WORKSPACE_ID/config" \ -H "Authorization: Bearer $OPENWORK_SERVER_TOKEN" ``` ## Audit log (recent) ```bash curl -s "$OPENWORK_SERVER_URL/workspace/$OPENWORK_WORKSPACE_ID/audit?limit=25" \ -H "Authorization: Bearer $OPENWORK_SERVER_TOKEN" ``` ## OpenCode engine checks ```bash opencode -p "ping" -f json -q opencode mcp list opencode mcp debug <name> ``` ## DB fallback (read-only) When the engine API is unavailable, you can inspect the SQLite db: ```bash sqlite3 ~/.opencode/opencode.db "select id, title, status from sessions order by updated_at desc limit 5;" sqlite3 ~/.opencode/opencode.db "select role, content from messages order by created_at desc limit 10;" ``` ## Notes - Audit logs are stored at `.opencode/openwork/audit.jsonl` in the workspace root. - OpenWork server writes only within approved workspace roots.
More from different-ai/openwork
- browser-setup-devtoolsGuide users through browser automation setup using Chrome DevTools MCP only. Use when the user asks to set up browser automation, Chrome DevTools MCP, browser MCP, or runs the browser-setup command.
- daytona-electron-testTest the real Electron app on Daytona: create sandbox, start services, connect via CDP, create workspaces, drive sessions, and verify settings. Use when the user says 'test on Daytona', 'run the app on Daytona', 'Daytona dry run', 'test Electron remotely', or 'reproduce on Daytona'.
- get-startedGuide users through the get started setup and Chrome DevTools demo.
- opencode-bridgeBridge between OpenWork UI and OpenCode runtime
- opencode-mirrorMaintain the local OpenCode mirror for self-reference
- opencode-primitivesReference OpenCode docs when implementing skills, plugins, MCPs, or config-driven behavior.
- openwork-coreCore context and guardrails for OpenWork native app
- openwork-orchestrator-npm-publish|
- run-evalsRun OpenWork UI evals on a Daytona sandbox or local Electron instance. Handles sandbox creation, service startup, and eval execution via CDP browser tools.
- shadcnManages shadcn components and projects — adding, searching, fixing, debugging, styling, and composing UI. Provides project context, component docs, and usage examples. Applies when working with shadcn/ui, component registries, presets, --preset codes, or any project with a components.json file. Also triggers for "shadcn init", "create an app with --preset", or "switch to --preset".