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/omarshahine/restaurant-cli/restaurant-snipegit clone --depth 1 https://github.com/omarshahine/restaurant-cliWrote 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/commands/omarshahine/restaurant-cli/restaurant-snipe)<a href="https://agentmods.dev/commands/omarshahine/restaurant-cli/restaurant-snipe"><img src="https://agentmods.dev/badge/commands/omarshahine/restaurant-cli/restaurant-snipe.svg" alt="Measured on agentmods" 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 | $0.00025 | $0.00552 |
| Opus 5 | $0.00013 | $0.00276 |
| Sonnet 5 | $0.00005 | $0.00110 |
| Haiku 4.5 | $0.00003 | $0.00055 |
Grade A, and why
restaurant-snipe 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 5d 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.
What it actually says
Restaurant Snipe
Off by default. Scheduled sniping requires the user to opt in by setting
RESTAURANT_CLI_ENABLE_SNIPE=1in their environment (it's an unattended booking). Ifrestaurant snipeerrors with "off by default", relay that to the user and let them enable it — do not set the env var yourself.restaurant snipe --dry-runpreviews without the flag.
The user wants to queue a sniped booking. Collect the following (via AskUserQuestion if any are missing):
- venue id — if the user gave a name, run
restaurant search "<name>"first and confirm the venue - party size
- date (YYYY-MM-DD)
- time (HH:mm, 24h)
- release-at — ISO8601 with timezone offset, e.g.
2026-04-30T10:00-04:00. This is when the at-job fires, not the reservation time. - provider — default: configured default; must have
snipecapability perrestaurant doctor. As of 2026-04, only Resy supports sniping.
Then run:
restaurant snipe --provider <id> --venue <id> --party <n> --date <d> --time <t> --release-at <iso>
The CLI will prompt for y/N confirmation unless you pass --yes. Confirm with the user in chat first, then pass --yes so there's no second prompt.
After queuing, tell the user:
- The job id and
runAttimestamp restaurant jobs listto inspect the queuerestaurant jobs cancel <job-id>to back outrestaurant jobs logs <job-id>to see the fire-time output (available after the job runs)
Under the hood, the CLI pipes the job to POSIX at and writes a bash wrapper that sources ~/.secrets.env at fire time so the booking command has access to RESY_AUTH_TOKEN.
Heads up — deferred credential access. A snipe is an unattended booking: the wrapper loads your provider tokens and fires the reservation later, with no further confirmation from you. Tell the user this when they queue one, and remind them they can back out any time with
restaurant jobs cancel <job-id>. Only queue snipes the user has explicitly authorized.
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.
- 5d ago First seen · 35 lines · 25 tokens per session scan A a58948027494
restaurant-snipe is a command published in the GitHub repository omarshahine/restaurant-cli (8 stars, last pushed 3d ago), licensed MIT. It adds 25 tokens to every session and 552 once invoked, about $0.0001 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
cli-anything-web
Generate a complete agent-native CLI for any web app by recording and analyzing network traffic via playwright-cli. Runs the full pipeline with site assessment, capture, implementation, testing, and verification.
cli-anything-web:refine
Refine an existing cli-anything-web CLI by recording additional traffic and expanding command coverage. Invokes the gap-analyzer skill as its first step, then implements missing endpoints.
act
Live agentic desktop + browser control via OAuth (no API key).
cli-anything-web:record
Record network traffic from a web app without generating a CLI. Useful for initial exploration or adding more coverage data.
naksha-browse
Browse any URL with browser vision — extract design patterns for research or inspect your own UI, saved to project memory.
design-compare
Capture two websites with Playwright and produce a structured side-by-side design comparison — layout, color, typography, navigation, and 'steal this' recommendations.