Plashboard Admin

Manage plashboard templates and autonomously convert natural-language dashboard requests into plashboard tool actions.

Security Scan

The skill's declared purpose, required artifacts, and runtime instructions are internally consistent: it's an instruction-only adapter that expects plashboard plugin tools and does not request extra credentials or install any code.

Purpose & Capability

Name/description (manage plashboard templates and convert NL requests) match the SKILL.md instructions, which mostly call plashboard_* plugin tools; no unrelated binaries, env vars, or installs are requested.

Instruction Scope

Instructions are narrowly focused on calling the listed plugin tools and include reasonable guardrails (don’t edit JSON files, avoid Docker/systemd). They also provide command-shortcuts; note the skill assumes the presence and correctness of many plashboard_* tools — if those tools perform unexpected actions the overall behavior could differ.

Install Mechanism

No install spec and no code files (instruction-only) — lowest-risk footprint; nothing will be written or downloaded by this skill itself.

Credentials

No environment variables, credentials, or config paths are requested; required artifacts are the named plugin tools, which is proportionate to the described functionality.

Persistence & Privilege

always is false and model invocation is allowed (normal). The skill does not request permanent presence or modify other skills; autonomous tool invocation is expected for this kind of admin skill.

Assessment

This skill itself is an instruction-only wrapper that delegates actions to a set of plashboard_* plugin tools. Before installing: 1) Verify the provenance and permissions of the plashboard_* tools the skill expects — those tools carry the real risk (they may access files, run commands, or modify systems). 2) Confirm guardrails are enforced by the runtime (the skill tells the model not to edit system files or run Docker, but the enforcement depends on the tooling). 3) Test in a non-production environment so you can observe what the underlying plashboard tools do when invoked. If you cannot verify or trust the plashboard plugins, avoid enabling autonomous invocation or do not install the skill.

Use this skill for plashboard runtime administration.

Use Cases

• Convert user requests like "I want dashboard X" into created/activated dashboards automatically.
• Create, update, copy, delete, and validate dashboard templates.
• Activate a template as the live dashboard source.
• Trigger immediate runs.
• Inspect runtime status and schedule health.
• Adjust display profile for layout budget checks.

Required Tooling

Always use plugin tools:

• plashboard_onboard
• plashboard_setup
• plashboard_exposure_guide
• plashboard_exposure_check
• plashboard_web_guide
• plashboard_doctor
• plashboard_permissions_fix
• plashboard_init
• plashboard_quickstart
• plashboard_template_create
• plashboard_template_update
• plashboard_template_list
• plashboard_template_activate
• plashboard_template_copy
• plashboard_template_delete
• plashboard_template_validate
• plashboard_run_now
• plashboard_status
• plashboard_display_profile_set

Guardrails

• Never edit /var/lib/openclaw/plash-data/dashboard.json directly.
• Never edit template/state/run JSON files directly.
• Never perform Docker, Tailscale, or systemd operations.
• Never ask the model to generate full dashboard structure when filling values.
• Do not tell end users to run slash commands when tool calls can do the action directly.

Intent Automation

• If the user asks for a new dashboard in natural language, call plashboard_onboard with:
  • description: user request rewritten as a concrete dashboard objective
  • force_quickstart: true
  • activate: true
  • run_now: true
• If onboarding returns readiness issues, call plashboard_web_guide and plashboard_exposure_guide, then present exact operator commands.
• If the user asks to modify an existing dashboard, call plashboard_template_list first, then update/copy/activate/run via tools.
• Prefer tool execution over conversational planning; only ask clarifying questions if the request is ambiguous.

Command Shortcuts

• /plashboard onboard <description> [local_url] [https_port] [repo_dir]
• /plashboard quickstart <description>
• /plashboard setup [openclaw [agent_id]|mock|command <fill_command>]
• /plashboard doctor [local_url] [https_port] [repo_dir]
• /plashboard fix-permissions [dashboard_output_path]
• /plashboard web-guide [local_url] [repo_dir]
• /plashboard expose-guide [local_url] [https_port]
• /plashboard expose-check [local_url] [https_port]
• /plashboard init
• /plashboard status
• /plashboard list
• /plashboard activate <template-id>
• /plashboard copy <source-template-id> <new-template-id> [new-name] [activate]
• /plashboard delete <template-id>
• /plashboard run <template-id>
• /plashboard set-display <width> <height> <safe_top> <safe_bottom>