Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Room Cli

ASecurity

Inspect and control Cloudroom Local and Cloud threads, projects, settings, plugins, and the cloud VM with room-cli. Use for Cloudroom CLI tasks, not official BB.

11 stars
0 votes
0 copies
0 views
Added 10/5/2026
ai-agentsshellgitbackend

Works with

terminalcli

Security Analysis

A100/100

Pro scans all 10 files and shows the line behind each finding

Scanned 10/6/2026

$npx -y skills add davidondrej/cloudroom-gui --skill room-cli --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Room Cli?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Room Cli
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/davidondrej-room-cli-cloudroom-gui/badge)](https://www.skillsdirectory.com/skills/davidondrej-room-cli-cloudroom-gui)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: room-cli
description: 'Inspect and control Cloudroom Local and Cloud threads, projects, settings, plugins, and the cloud VM with room-cli. Use for Cloudroom CLI tasks, not official BB.'
---

# Cloudroom CLI

Use `room-cli` for Cloudroom, not official BB's `bb`. Never inspect, message, or launch other threads unless the user asks.

## Check context first

```sh
room-cli status --json
```

Confirm the server URL and project/thread before acting. Standalone `room-cli` targets the installed Cloudroom backend at `http://127.0.0.1:39886`, not the cloud VM. Official BB remains separate.

Cloudroom supplies `ROOM_SERVER_URL`, `ROOM_HOST_DAEMON_PORT`, `ROOM_CLI`, `ROOM_DATA_DIR`, and thread context to Local agents and thread-scoped terminals automatically. Machine-only and environment-only terminals have no current thread. `room-cli` ignores the corresponding `BB_*` settings. Use `ROOM_*` overrides only for an intentional target change; never reuse thread IDs from another instance. `"$ROOM_CLI"` selects the current runtime's executable directly.

## Common commands

```sh
room-cli project list --json
room-cli thread list --project PROJECT --json
room-cli thread show THREAD --json
room-cli thread log THREAD
room-cli thread output THREAD
room-cli thread tell THREAD "Continue with this task"
room-cli thread stop THREAD
room-cli thread update THREAD --title "New title"
room-cli thread archive THREAD
```

"Open" or "active" threads means threads with `archivedAt: null`. An archived thread is never open or active, whatever its `status` says. `thread list` shows open threads only; `--archived` shows only archived ones, and `--include-archived` shows both.

`--self` uses `ROOM_THREAD_ID` where supported. Thread IDs select Local or Cloud execution; connection settings still point to the same GUI backend.

## Use Cloudroom from a terminal

`room-cli thread chat THREAD` turns any terminal pane (Herdr, tmux, plain shell) into a chat with a Local or Cloud thread: live output, messages, `/approve`, `/deny`, `/stop`, and `/exit`. Start a new agent in the pane with:

```sh
room-cli thread chat "$(room-cli thread spawn --project PROJECT --prompt "Task" --json | jq -r .id)"
```

Inside Herdr, `chat` shows the thread as `cloudroom` with its idle, working, or blocked state, and Herdr reopens the chat after a restart. Answer agent questions in the app or with `room-cli thread interactions`.

## Cloud threads

```sh
room-cli cloud status --json
room-cli thread spawn --project PROJECT --machine cloud --provider codex \
  --model MODEL --request-id UNIQUE_ID --prompt "Task"
```

- Supports Codex and Pi, inspection, plain-text follow-ups, stop, title changes, pinning, and archive/restore.
- `tell` queues Cloud messages automatically. Local messages still steer by default. Explicit unsupported modes fail; no fallback to Local.
- Stop pauses the Cloud queue. To resume, list it with `room-cli thread queue list THREAD`, then send its first message with `room-cli thread queue send THREAD MESSAGE_ID`.
- Read `room-cli cloud status --json` before optional Cloud operations. Steering, compaction, attachments, message editing (`rewind`), and queue editing/cancellation depend on the connected core and harness capabilities. Forks and scheduling are not enabled. The model is fixed per Cloud thread; change effort with `thread update THREAD --reasoning-level LEVEL` and read both with `thread show`. `--parent-thread` with a Cloud parent starts a linked child in that parent's sandbox. A Local parent's Cloud child (`--parent-self --machine cloud`) gets its own sandbox and messages the parent when it finishes; it starts from GitHub, so push first and pass `--base-branch BRANCH`. A capability flag is not proof that an operation completed.
- Retry an unconfirmed start with the same `--request-id`. For a rejected start, use `room-cli cloud retry-start THREAD`. Do not blindly resend uncertain work.
- To fix or set up the VM or a Cloud thread's sandbox, or move files, configs, and logins to it, use `room-cli vm run|pull|push`, with `--thread ID` for a sandbox (see the command index).
- Cloud VM agents do not get this CLI. They reach this Mac with `cloudroom mac run|pull|push`, only while "Let cloud agents access this computer" is on. They can rename, archive, or change the effort of their own thread with `cloudroom thread update --self --title TITLE`, `cloudroom thread archive --self`, and `cloudroom thread update --self --reasoning-level LEVEL`, start children in their own sandbox with `cloudroom thread spawn`, and stop or archive them with `cloudroom thread stop|archive CHILD_ID`. These commands control Cloud threads through the GUI backend while it is running.

## Verify results and Local setup

`thread wait --status idle` only checks a status. Queued work may still be waiting. Check the queue and correlate the requested task with its actual output; after compaction, verify a follow-up completes, and after an attachment upload, verify the agent can read its content.

Before a Local launch, inspect `room-cli machine provider-cli status MACHINE --json`. A cached model list does not prove the executable is installed or logged in. If Claude is installed but reports `Not logged in`, ask the user to run `claude` and `/login`; do not collect credentials or install another copy.

## References

Use `room-cli --help`, `room-cli GROUP --help`, or `room-cli guide CHAPTER` for current options. Read only the relevant reference:

- [Command index](references/command-index.md): all core command paths.
- [Thread creation](references/thread-creation.md): projects, machines, Local launches and forks.
- [Thread operation](references/thread-operation.md): inspection, messaging, queues, files, terminals.
- [Failure recovery](references/failure-recovery.md): failed/stopped Local threads and provider controls.
- [Configuration](references/configuration.md) and [app settings](references/app-settings.md): settings and environment setup.
- [Plugins](references/plugins.md): plugin management; each plugin owns its command instructions.
- [Theme commands](references/theme-commands.md) and [theming](references/theming.md): appearance.

Resolve IDs before mutation. Prefer JSON for scripts. Confirm actual results and report the ID the user needs. Never claim a queued message has completed.

Attribution

davidondrejdavidondrej
View sourceSee grades on GitHubMore from davidondrej →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698621 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →