Tables Capacity logo

Tables Capacity

Organization
clay-run
tables-capacity

Clay tables — check the row-capacity ceiling when imports stall, before assuming a run/config/enrichment problem. Use when the user asks "why aren't new rows being added?", "why is the import stuck?", or "is this table full?".

Overview

Publisherclay-run
Repositoryagent-plugins
Skill nametables-capacity
Stars
145
Forks
24
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by clay-run on GitHub. Read the source before you install it.

Installation

Install the Tables Capacity AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/clay-run/agent-plugins/tree/main/clay/skills/tables-capacity
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/clay-run/agent-plugins.git /tmp/agent-plugins
mkdir -p .claude/skills
cp -r /tmp/agent-plugins/clay/skills/tables-capacity .claude/skills/tables-capacity
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Tables Capacity in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Tables Capacity on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Tables Capacity is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Capacity check

Use when: "why aren't new rows being added?", "why is the import stuck?", "is this table full?", or any question about rows failing to appear. Check capacity before assuming a run, config, or enrichment problem — a full table silently stops accepting rows, which looks like a broken import.

Scope: regular tables only. Archive tables are not subject to this ceiling — if the user points at an archive (t_... that came from another table's archive.tableId), say the check doesn't apply and redirect to the parent table.

The rule

A Clay table stops accepting new rows once it reaches its row ceiling. The ceiling is 50,000 by default, but it is plan- and table-dependent — plans can set it lower, and per-table overrides can raise it — and the CLI does not expose the effective value. The same ceiling logic applies to:

  • rowCount — the table's own row total (clay tables get). At the ceiling, the table won't take new rows.
  • each source's numSourceRecords — a source feeding the table, read from the table's source columns (clay tables columns list → sources[] on each source column). A capped source can't push more records in, even if the table itself has room.

So the table is constrained if rowCount OR any source's numSourceRecords is at its ceiling. Evaluate every source, not just the largest.

Procedure

  1. Resolve the table as in the tables entry-point skill ("Finding a table").
  2. Run clay tables get for rowCount, and clay tables columns list for the source counts (shapes in each command's --help). Two commands — no row scan needed.
  3. Read the numbers against the default ceiling (source counts live on source columns), with two separate commands.

Read the table's own row total:

bash
clay tables get <tableId> | jq .rowCount

Read each source's count:

bash
clay tables columns list <tableId> | jq '[ .data[] | select(.type == "source") | .sources[] | {name, numSourceRecords} ]'

Then compare rowCount and every source's numSourceRecords against the default ceiling (50,000) yourself — the table is constrained if any of them is at or above it.

Interpreting the numbers

  • At or above 50,000 (rowCount or any source) → the table has hit the default ceiling and new rows are being rejected. This is the answer.
  • Stuck at a suspiciously round number below 50k (rowCount pinned at e.g. exactly 25,000 while an import is supposedly running) → likely at a lower plan ceiling. The CLI can't confirm the effective limit — report the pattern and say so, rather than ruling capacity out.
  • Below the ceiling and moving → capacity is not the blocker; look elsewhere.

Proximity is not urgency. Don't classify "close to the ceiling" as a problem by itself — whether 47k matters depends entirely on how fast the table is filling. When the user needs a runway estimate, measure the fill rate: read rowCount, wait a few minutes, read it again, and report time to ceiling (headroom / rate), not percent-of-ceiling. If the counts aren't moving, there is no urgency to report — just state the headroom.

Answering

Lead with the verdict, then the number that drives it. Name whether it's the table's own rowCount or a specific source.

FULL — table's own rows:

People — FULL (rowCount 50,000, the default ceiling).
The table has hit its row ceiling, so new rows are being rejected.
That's why the import isn't adding records — it's not a run or config issue.
Fix: split into another table, or remove rows to free capacity.

FULL — a source is capped:

Leads — FULL via source "Salesforce Import" (numSourceRecords 50,000).
rowCount is 41,200, so the table itself has room, but that source can't push more
records in. New rows from Salesforce won't appear until the source is under its cap.

Not full — with headroom and, when asked, runway:

Accounts — not at the ceiling (rowCount 47,300; 2,700 rows of headroom against the
default 50k). Measured fill rate ~600 rows/hour over the last 5 minutes → roughly
4-5 hours to the ceiling at the current pace. Worth splitting or pruning today.
Contacts — not at the ceiling (rowCount 12,400; ~37,600 rows of headroom). Capacity
isn't the problem here.

When capacity is ruled out, move on to whatever the real question was (an errored source, an enrichment that didn't run, etc.). If the user came in with "rows aren't appearing," hand off to /tables-error-sweep or /tables-trace from here.

Notes

  • No source columns → the check reduces to rowCount alone (the jq handles this).
  • Multiple sources → report each source's count; the constraint is whichever one (or the table) is at its ceiling, so don't stop at the first.
  • The ceiling is product behavior, not something the CLI returns — the commands only give you the counts, and 50,000 is the default, not a guarantee. Rejection behavior (counts pinned while an import runs) is the ground truth.

Frequently asked questions

What does the Tables Capacity AI skill do?

Clay tables — check the row-capacity ceiling when imports stall, before assuming a run/config/enrichment problem. Use when the user asks "why aren't new rows being added?", "why is the import stuck?", or "is this table full?".

Why use Tables Capacity on TypingMind?

Because you install it once and use it with any model. Tables Capacity is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Tables Capacity in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/clay-run/agent-plugins/tree/main/clay/skills/tables-capacity. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Tables Capacity?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Tables Capacity?

As many as you like. As long as a model supports skills, you can use Tables Capacity with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Tables Capacity AI skill free?

It is published on GitHub by clay-run. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇