>- Optimize table turns, manage reservations, and reduce wait times to maximize revenue per seat. Handles booking intake, table assignment strategy, waitlist management, and provides actionable insights for seating efficiency.
npx skills add https://github.com/cosmicstack-labs/mercury-agent-skills --skill table-manager
For a restaurant, every empty seat is revenue walking out the door. A table that sits empty for 30 minutes during peak dinner service costs more in lost margin than most marketing campaigns can recover. Table Manager is about maximizing revenue per available seat hour (RevPASH) — the restaurant equivalent of hotel RevPAR.
Ask the user for:
If they don't know average turn times:
> "Let me calculate. What time was the first party seated at each table, and what time did the last party leave? I'll compute actual turn times from your service logs."
Revenue Per Available Seat Hour (RevPASH) = Total Revenue ÷ (Total Seats × Hours Open)
Seat Utilization % = Actual Covers ÷ (Total Seats × Turns Possible)
Theoretical Max Revenue = Seats × Hours × Max Turns × Avg Check
Revenue Gap = Theoretical Max − Actual Revenue
Example:
> "Your restaurant has 40 seats open for 5 dinner hours. With 90-minute average turns and $45 avg check, your theoretical max is 133 covers / $5,985. Actual was 98 covers / $4,410. Gap: $1,575 in unrealized revenue. Every 10-minute reduction in turn time recovers $350/night."
Booking intake rules:
No-show management:
> "Your no-show rate is 12% this month. At $45 avg check per cover, that's $540 in lost revenue per 100 reservations. Consider: (1) SMS reminders 4 hours before, (2) credit card on file for 5+ parties, (3) waitlist priority for no-shows' next booking."
When seating a party, recommend optimal table:
| Party Size | Best Fit | Why |
|-----------|----------|-----|
| 2 ppl | 2-top (not 4-top) | A 2-top at a 4-top wastes 2 seats per turn |
| 3 ppl | 4-top, pushed to one side | Leaves room for a 2-top on the other half if possible |
| 4 ppl | 4-top | Perfect fit |
| 5-6 ppl | 6-top or push two 4-tops | Don't use a 6-top for a 4-top — always match party to table size |
| 7+ ppl | Multiple tables or private area | Requires coordination with kitchen timing |
Smart seating rules:
When the restaurant is at capacity:
Waitlist dashboard:
🪑 WAITLIST STATUS — Friday 7:15pm
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Tables Available: 0 (all 12 tables seated)
Expected Free: 2 tables in ~15 min (check-present stage)
WAITING: 7 parties (estimated 25 min avg wait)
1. Smith (2) — waiting 12 min — will get next free 2-top
2. Jones (4) — waiting 8 min
3. Patel (6) — waiting 5 min — waiting for 6-top
4. Lee (2) — just arrived
...
🔄 TABLE TRACKER
Table 3 (2-top): Check presented — clearing in ~3 min
Table 7 (4-top): Entrees served — ~25 min remaining
Table 12 (6-top): Just seated — ~75 min remaining
Analyze party size distribution vs. table configuration:
> "Your party data shows 45% are 2-tops, but only 25% of your tables are 2-tops. You're seating 2-tops at 4-tops 40% of the time, wasting 2 seats per turn. Converting two 4-tops into 2-tops would recover an estimated 8-12 additional covers per evening — worth ~$450/night or ~$14,000/month."
When things go wrong (long wait, table mix-up):
| Phrase | What Happens |
|--------|-------------|
| "How's the floor looking?" | Shows current table status and waitlist snapshot |
| "We have a [size] party walking in" | Recommends best table assignment |
| "What's the wait time?" | Calculates estimated wait based on table turns ahead |
| "Optimize my table layout" | Analyzes table mix vs. party size data, recommends reconfiguration |
| "How many covers can we do tonight?" | Calculates capacity based on reservations + walk-in allocation |
| "Track my turn times" | Monitors and reports average turn time by meal period |
| "Analyze my no-show rate" | Reports no-show % and recommends prevention strategies |
| "Add [name/party] to the waitlist" | Logs party, provides quote, manages queue position |
*"Your tables are your most valuable real estate. Every empty seat has a cost."*
Integration with protocols.io API for managing scientific protocols. This skill should be used when working with protocols.io to search, create, update, or publish protocols; manage protocol steps and materials; handle discussions and comments; organize workspaces; upload and manage files; or integrate protocols.io functionality into workflows. Applicable for protocol discovery, collaborative protocol development, experiment tracking, lab protocol management, and scientific documentation.
Analyzes job descriptions and generates tailored resumes that highlight relevant experience, skills, and achievements to maximize interview chances
Generate Excalidraw diagrams from natural language descriptions. Use when asked to "create a diagram", "make a flowchart", "visualize a process", "draw a system architecture", "create a mind map", or "generate an Excalidraw file". Supports flowcharts, relationship diagrams, mind maps, and system architecture diagrams. Outputs .excalidraw JSON files that can be opened directly in Excalidraw.
Build and distribute Expo development clients locally or via TestFlight
Use when you have a written implementation plan to execute in a separate session with review checkpoints
Data structure for annotated matrices in single-cell analysis. Use when working with .h5ad files or integrating with the scverse ecosystem. This is the data format skill—for analysis workflows use scanpy; for probabilistic models use scvi-tools; for population-scale queries use cellxgene-census.
Benchling R&D platform integration. Access registry (DNA, proteins), inventory, ELN entries, workflows via API, build Benchling Apps, query Data Warehouse, for lab data management automation.
Comprehensive molecular biology toolkit. Use for sequence manipulation, file parsing (FASTA/GenBank/PDB), phylogenetics, and programmatic NCBI/PubMed access (Bio.Entrez). Best for batch processing, custom bioinformatics pipelines, BLAST automation. For quick lookups use gget; for multi-service integration use bioservices.
Take cosmicstack-labs/table-manager from the repository into ~/.claude/skills for personal
use, or into .claude/skills inside a project.
The agent identifies a skill by the name field in its header. Two skills with the
same name cannot sit side by side — one of them will be ignored.