Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This collection of tools assists outbound sales representatives in identifying decision-makers and launching personalized outreach. It streamlines the transition from a target company domain to a live campaign by automating website scraping, lead extraction, and message personalization.
By integrating with Prospeo for lead pulling and Smartlead for campaign deployment, the system removes the manual friction of building prospect lists and drafting individual emails.
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
Point Claude Code at all signal playbooks
Installs: All 19 signal playbooks under `skills/playbooks`; the skillPaths list is the supported bounded inventory of those playbooks. · Claude Code
Before you start
Claude Code installed
Repository already cloned
Run from: Cloned repository root
In Claude Code, point the agent at the cloned repository's `skills/playbooks` directory. These playbooks live one level deeper than the other skills, so the repository root alone does not establish their discovery.
Path to a lead CSV containing at least 100 rows and the documented minimum columns
Optional path to a client-profile YAML
Output path for the Markdown scorecard
Check the repository for this skill’s setup.
Use the list-quality-scorecard skill to grade `[path/to/leads.csv]`, which must contain at least 100 rows, and write the scorecard to `[path/to/scorecard.md]`. The CSV contains `email`, `first_name`, `last_name`, `job_title` or `title`, and `company_name` or `company`. If available, also compare it with `[path/to/client-profile.yaml]`. Report the letter grade, eight dimension scores, top five issues, and pre-send checklist; do not upload or send anything.
Use the deliverability-test-public skill to compare SMTP, Gmail/G Suite, and Outlook/Office365 performance across my Smartlead account for the last 7 days. Report sent volume, replies, bounces, reply rate, and bounce rate by inbox type, and flag small samples as inconclusive. Do not change campaigns or inbox settings.
Explicit ICP headcount, country, and title criteria
Confirmation that the source company and its employees may be added to the client's do-not-contact list
Use the playbook-linkedin-engagement skill for a one-post pilot. The source account is `[LinkedIn company URL]`, the client's domain is `[client.example]`, and the confirmed ICP is `[headcount, countries, and titles]`. Do not generate an `engagement_line`. Resolve the source to a bare domain first, stop if it cannot be resolved or if the client's own domain appears, and ask me to confirm that the source company and its employees will enter the client's do-not-contact list before any paid call. Return only the pilot prospect rows and evidence URLs; do not scale beyond one post.
Claude