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/lookatitude/guild/dashboardgit clone --depth 1 https://github.com/lookatitude/guildWhat 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 | $0.00068 | $0.00994 |
| Opus 5 | $0.00034 | $0.00497 |
| Sonnet 5 | $0.00014 | $0.00199 |
| Haiku 4.5 | $0.00007 | $0.00099 |
Grade A, and why
dashboard 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 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.
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 — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/guild:dashboard — visual dashboard over this project
Launches the benchmark factory's web UI against the live project
(--project-root): imported runs, the knowledge-graph explorer, and the wiki
browser. Server binds 127.0.0.1 only; project routes are read-only.
Initiative: guild-dashboard (WI-3). Contract pinned in
.guild/spec/v2-gap-closure.md §"The cross-repo dashboard contract".
Args & local flags
- Args: — (no positional)
--port N— preferred port (default 3055; next free port used when bound).--no-import— skip the bulk lifecycle import of.guild/runs/*.--install— pre-approve the network leg (git clone / npm ci) when no benchmark checkout exists. Without it the launcher prints the exact commands and exits (REQUIRED-INSTALL) — it never touches the network.--dry-run— print the resolved plan (checkout, port, run dirs, serve command); execute nothing.--stop— managed shutdown: read the PID record (.guild/cache/dashboard.json), SIGTERM the server's process group, remove the record. Stale records (dead pid) are cleaned up and reported.
Gates
- I (interactive) on REQUIRED-INSTALL: clone/install is network +
always-ask class — the skill surfaces the exact commands and asks before
re-running with
--install. Everything else is localhost-only and ungated.
Output
Prints the dashboard URL (http://127.0.0.1:<port>/), the imported-run
summary (per-run failures non-fatal), and how to stop the server:
/guild:dashboard --stop (managed — the spawn is tracked by a durable PID
record at .guild/cache/dashboard.json with server output in
.guild/cache/dashboard.log; never a fire-and-forget daemon). Re-invoking
while the server is alive reuses it (no second server). No .guild/ data
writes (only .guild/cache/: PID record + log always; benchmark/ under a
confirmed --install).
Capability candidates (F7)
Alongside the URL, print the capability-candidate surface — the same read-only
block /guild:status shows:
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 · 88 lines · 68 tokens per session scan A b03f5f74e214
dashboard is a command published in the GitHub repository lookatitude/guild (7 stars, last pushed 2d ago), licensed MIT. It adds 68 tokens to every session and 994 once invoked, about $0.0003 per session on Opus 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 2026-08-31.
Other commands, from other repositories
recovery-crawl
Screaming Frog crawl with technical SEO issue classification including Shopware-specific patterns.
tech-product-search
Search the four tier-1 Israeli tech retailers — Ivory, KSP, Bug, TMS — for a consumer-tech product (laptop, phone, peripheral, accessory, charger, cable, monitor, TV, headphones, etc.) and return prices ranked low-to-high. Handles Hebrew-term resolution and falls back to Playwright when Tavily/WebFetch is blocked.
design-review
description: Run the 7-phase design review (WCAG AA, responsive, interaction) on a page or URL argument-hint: [url-or-file] [optional focus, e.g. "mobile nav" or "checkout flow"].
e2e
使用 Playwright 对 Web UI 进行端到端测试（支持视频录制、Trace 录制、控制台/网络日志捕获）.
auto-browse
Auto-browse — learn, optimize, and graduate browser operations or web data-mining workflows.
swarm-planner
Generates a structured execution plan with dependency tracking using the Swarms-inspired planning methodology.