Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/claudetool/fleetline/statusline-setupgit clone --depth 1 https://github.com/ClaudeTool/fleetlineWhat 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 | $0.00027 | $0.02208 |
| Opus 5 | $0.00014 | $0.01104 |
| Sonnet 5 | $0.00005 | $0.00442 |
| Haiku 4.5 | $0.00003 | $0.00221 |
Grade B, and why
statusline-setup scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- Home settings file: !`cat "$HOME/.claude/settings.json" 2>/dev/null || echo "(không tồn tại)"` How it starts
The opening of the file, as written. The whole thing — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Context
- Home settings file: !
cat "$HOME/.claude/settings.json" 2>/dev/null || echo "(không tồn tại)" - Existing config file (nếu đã setup trước đó): !
cat "$HOME/.claude/statusline-fleetline.config.json" 2>/dev/null || echo "(chưa có)"
Your task
You are setting up the fleetline plugin for the user. Follow these steps in order. Do not skip the backup step under any circumstances.
-
Ask with
AskUserQuestion(one call, up to 4 questions):- Layout: "Minimal" (1 line), "Hardened" (2 lines, matches the classic layout — recommend this as default), or "Power-user" (3 lines, adds cost/session/PR/thinking/bg-agents/cache-TTL when present).
- ASCII-safe mode: on/off. Recommend OFF unless the user says their terminal/font doesn't render Unicode or Nerd Font glyphs well.
- Background-agent segment: on/off. Explain the real cost/benefit honestly: it calls
claude agents --jsonto show how many background agents are running, but each call spawns a full CLI process (~0.4s observed) — throttled to once every N seconds (ask for N, default 60, minimum 5), not on every render. Recommend OFF as the default; only worth it if the user regularly runs multiple background/dispatched sessions. - If background-agent segment is ON: ask a follow-up whether to also enable the hook-based approximate counter (fills the gap between polls with a
~Nestimate; requires restarting Claude Code for the new hooks to load per Claude Code's hook-loading model, and can drift high if an agent is killed abnormally — self-corrects at the next poll). Recommend OFF unless the user wants faster-updating estimates and is fine with the restart.
-
Ask separately whether to also enable the fleet-view row renderer (
subagentStatusLine— a DIFFERENT settings key from the main statusline, rendering rows in Claude Code's multi-agent/fleet view). Be upfront: this is the least-verified part of the plugin — it was built from documented field names but never tested against a real live multi-agent payload, only synthetic data matching the documented shape. Default recommendation: skip it unless the user specifically wants to try it and is fine with rough edges.
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.
- 2d ago First seen · 94 lines · 27 tokens per session scan B cc1153a6f907
statusline-setup is a command published in the GitHub repository ClaudeTool/fleetline (4 stars, last pushed 6d ago), licensed MIT. It adds 27 tokens to every session and 2,208 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
test-tdd
Run when user calls /test-tdd. Scans modified files, locates their corresponding unit/integration test suites, and runs them.
skene-status
Check implementation status from skene-context/engine.yaml. Validates features, action vs code-only mode, and that trigger/function SQL exists in migrations; detail shows latest matching file and (+N) if duplicates.
plan
Execute the implementation planning workflow using the plan template to generate design artifacts.
pi-code-review
Adversarial code review via Codex — break confidence in changes, not validate them.
setup
Fetch Kobiton credentials from the authenticated MCP server and write them to /.kobiton/.credentials.
setup
Install the claude-pace statusline bundled with this plugin.