clay-playbooks

Generates copy-ready personalization lines from 19 buying signals via Clay tables and workflows.

Updated May 28, 2025
One-click install
npx skills add https://github.com/meloShaya/insteltech --skill clay-playbooks-meloshaya
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: clay-playbooks
Source: https://github.com/meloShaya/insteltech/tree/main/crm/library/skills/playbooks/clay-playbooks
Command: npx skills add https://github.com/meloShaya/insteltech --skill clay-playbooks-meloshaya

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Cold email personalization requires turning buying signals (new role, funding round, hiring surge, pricing page) into one accurate sentence per prospect, and doing it cheaply at scale without hallucinated facts or runaway enrichment costs. ## Core Features & Use Cases - 19 signal playbooks: Each playbook converts one buying signal (person-level, company-level, website-derived, list-shaping, or copy-generation) into a single output field like new_in_role_line or funding_line. - Three run modes per playbook: Run as a Claude skill on ad-hoc CSVs, as a Clay table built through a browser harness, or as a Clay workflow built from the clay CLI. - Shared harnesses: clay-table-harness.md and clay-cli-harness.md document the generic build procedures, including the AI-column reasoning-level fix, formula placeholder IDs, credit gates, and CLI wiring pitfalls. - Use Case: Given a 1,000-row prospect list, run the fundraising playbook to produce a funding_line per row, gate paid enrichment behind email validation, smoke test 5 rows, then push the field to your sequencer as a custom variable. ## Quick Start Ask the assistant to run the new-in-role playbook on your prospect CSV and produce a personalization line for each row.

Frequently Asked Questions about clay-playbooks

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I personalize cold emails with buying signals in Clay?

Pick the playbook matching your signal, build its Clay table columns or CLI workflow, and bind the AI column to the locked prompt. Each playbook outputs one field like `funding_line` that slots into a single sentence of your email.

What is the difference between a Clay table build and a Clay workflow build?

Tables are built in the browser because the clay CLI's tables commands are read-only. Workflows can be created end to end from the terminal using `clay workflows create`, `nodes create`, and `publish`, making them scriptable and version-controllable.

Why is my Clay AI column returning empty output on every row?

The default reasoning effort makes the model spend its token budget on reasoning and return empty content. Set `reasoningLevel` to the lowest value in the AI column bindings; do not fix it by raising `reasoningBudget`, which costs roughly 35x more per row.

Can the clay CLI create tables and columns?

No. The CLI's `tables` command group is read-only (`list`, `get`, `columns`, `rows`, `query`). Tables must be built in the browser, while workflows support full create, publish, and run operations from the CLI.

How do I control enrichment costs when running signal playbooks?

Gate every paid column with a run condition requiring a passed free upstream check and a valid email, and skip rows where the output is already filled. Source rows are tagged FREE, CHEAP, METERED, or EXPENSIVE relative to a 1,000-row list.

Are the Clay table and workflow recipes verified to work?

No. The `clay-table.md` and `clay-workflow.md` files are documented recipes written against the real Clay action catalog and CLI surface, but they have not been built in a live workspace. Build one, run it on 5 rows, and fix the file before scaling.