Flightdeck
A live agent dashboard pane: context, cost, permission checks and subagents.
Illustration. Actual results depend on the mod, terminal and environment.
Install
/plugin marketplace add scasella/claude-flightdeck
/plugin install flightdeck@claude-flightdeck
/reload-pluginsRun these commands one at a time in a Claude Code session.
claude plugin marketplace add scasella/claude-flightdeck
claude plugin install flightdeck@claude-flightdeckThen run /reload-plugins in your session, or start a new session.
Please install the Claude Code mod "Flightdeck" (https://github.com/scasella/claude-flightdeck). Before installing, read the Security and Requirements sections of its README and tell me what it reads and writes, which commands it runs, whether it uses the network, and which Claude Code version, OS and terminal it needs. If everything looks fine, run claude plugin marketplace add scasella/claude-flightdeck and claude plugin install flightdeck@claude-flightdeck, then remind me to run /reload-plugins in my session.Send this to Claude Code so it reads the author’s instructions before installing.
git clone https://github.com/scasella/claude-flightdeck
cd claude-flightdeck
claude --plugin-dir .Clone the repository and try it in a local session with --plugin-dir.
What it accesses
Based on the author’s documentation, checked when listed; use claude plugin validate to check for yourself.
Reads
YesReads tool inputs, permission verdicts, subagent events, the first 70 prompt characters and usage.
Writes files
NoneRuns commands
NoneNetwork
NoneIntroduction
Flightdeck puts a live agent dashboard inside Claude Code, and every number comes from a real session event. /flightdeck opens it; in fullscreen it docks beside the transcript, with two columns from 110 columns wide, and on the main screen it shows an 8-row summary above the prompt.
- main: model, effort, permission mode, a context gauge with compactions, cost and plan rate limits
- gate: one cell per permission check, green auto-allowed, blue decided by the classifier or you, red denied, with drill-downs by file, shell and other tools
- agents: subagents as side-by-side cards, switching to swimlanes on one time axis when they no longer fit; plus an architect consult timeline, a last-turn receipt and a log
- An optional status line shows context, running agents, consults and denials
The author states it only watches and never denies or rewrites a call. It makes no network requests, runs no processes, reads and writes no files, stores nothing across sessions and calls no model; prompts are kept to their first 70 characters for the log. Requires Claude Code 2.1.287+; VS Code and mobile cannot animate. Not run-tested by this site; the preview is illustrative.
Who it is for
- People running several subagents who want to see who is doing what
- People watching context, cost and permission checks
Requirements
- Claude Code ≥ 2.1.287
- Docks beside the transcript in fullscreen; two columns from 110 columns.
- VS Code and mobile show it without animation.