Borrowing it
Nothing to install: this file belongs to brizzai/fleet. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/brizzai/fleet/master/.claude/skills/drive-fleet/SKILL.mdgit clone --depth 1 https://github.com/brizzai/fleetWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/brizzai/fleet/drive-fleet)<a href="https://agentmods.dev/skills/brizzai/fleet/drive-fleet"><img src="https://agentmods.dev/badge/skills/brizzai/fleet/drive-fleet/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/brizzai/fleet/drive-fleet"><img src="https://agentmods.dev/badge/skills/brizzai/fleet/drive-fleet.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00156 | $0.01881 |
| Opus 5.5 | $0.00062 | $0.00752 |
| Sonnet 5.5 | $0.00031 | $0.00376 |
| Haiku 4.5 | $0.00016 | $0.00188 |
Grade A, and why
drive-fleet scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Drive the fleet TUI
make build proves a UI change compiles. It says nothing about whether the
dialog fits, the column aligns, or the footer is still on screen. This skill
closes that gap: boot a seeded fleet in a sandbox, press keys, read the screen.
The whole surface is demo/drive.sh. It drives a throwaway fleet under
/private/tmp/fleet-drive on its own tmux server — never the user's real
sessions. That holds when you are yourself running inside a tmux pane, which
you usually are: the script clears $TMUX before every tmux call, because
$TMUX names a server and outranks TMUX_TMPDIR, and down verifies the
socket it is about to kill lives under the sandbox.
Step 1: Bring it up
demo/drive.sh up # builds if needed, seeds, launches at 120x40
demo/drive.sh up 80x24 # or start at a size you want to test
First run takes a few seconds (Go build + git init on three repos). After
that up reuses the sandbox. If it is already running, up relaunches.
Step 2: Drive it
demo/drive.sh key S # key names: S, Down, Enter, Escape, C-k
demo/drive.sh key Down Down Enter # several keys in order
demo/drive.sh type "my branch name" # literal text — never key-name parsed
demo/drive.sh size 80x24 # resize; the pane really is 24 rows
key and type are different tools and neither substitutes for the other.
key hands its arguments to tmux's key-name parser, so Down means the arrow.
type sends literal characters, so Down means the word.
Step 3: Read the screen
demo/drive.sh snap # settled plain-text grid
demo/drive.sh snap -e # keep ANSI, to check colors
demo/drive.sh snap -w 'Appearance' # settle AND assert, in one call
demo/drive.sh png /tmp/frame.png # image of that same frame
Prefer -w whenever you know what should appear. Without it you race the
dialog's first paint; with it, the wait and the assertion are one atomic step
and a failure is loud instead of a confusing empty grep.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 167 lines · 156 tokens per session scan A a7571b342cff
drive-fleet is a skill published in the GitHub repository brizzai/fleet (59 stars, last pushed 9d ago), licensed Apache-2.0. It adds 156 tokens to every session and 1,881 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-07.
Other skills, from other repositories
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…
test-master
Generates test files, creates mocking strategies, analyzes code coverage, designs test architectures, and produces test plans and defect reports across functional, performance, and security testing disciplines. Use when writing unit tests, integration tests, or E2E tests; creating test strategies or automation…
eval-dataset-design
A guide to creating reliable test sets for AI agents. Each evaluation task includes the request, the simulated user's behaviour, a reset starting state, and a success check that can be independently verified.
dogfood
Systematically explore and test a mobile app on iOS/Android with agent-device to find bugs, UX issues, and other problems. Use when asked to dogfood, QA, exploratory test, find issues, bug hunt, or test this app on mobile.
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
critique
Adversarial design critique of the current work. Render it, look at it, and argue for rejection. Run after the gates are green, never instead of them.