Stage: Signal activation

Create Contact Signals

Use this skill to run contact-level signal research using the Saber CLI.

Saber CLI check

Before doing anything else, check if the Saber CLI is installed by running saber --help.

If the CLI is available: proceed with the full workflow below.

If not installed: inform the user that this skill requires the Saber CLI to execute signals (available at saber.app), then offer two options:

1. Install the CLI — once installed, come back and restart this skill
2. Continue without the CLI — document the approved signal questions and contact LinkedIn URLs so the user can run them later via the CLI; walk through the workflow conceptually so they know what to expect

Prerequisites

• Approved contact signal definitions are available in conversation context (run signal-discovery first if not)
• Saber CLI is available (saber --help works)

Before running signals

Before executing any saber signal or saber subscription command that will consume credits:

1. Run saber credits and show the user their current balance.
2. Tell the user how many signals will be run (number of contacts × number of questions).
3. Ask the user to confirm they want to proceed before issuing the command.

Two modes

Mode A — Spot-check a specific contact

Use saber signal with a LinkedIn profile URL to run a question against a specific contact:

saber signal --profile <linkedin-url> --question "<signal question>"

This is synchronous by default — it waits for the result and prints it.

# Example
saber signal --profile https://linkedin.com/in/janedoe --question "Is this person posting about employee retention challenges?"

Use --no-wait to fire multiple signals without waiting:

saber signal --profile <linkedin-url> --question "<question>" --no-wait
# Returns a signal ID; retrieve result later with:
saber signal get <signalId>

Mode B — Run signals across a large contact list (signal subscriptions)

For running signals across a large contact list, use signal subscriptions in the Saber dashboard or via the API. The CLI currently supports per-contact signals only.

Workflow (spot-check mode)

Step 1 — Pick contacts to check

Ask the user which contacts they want to prioritise. You'll need LinkedIn profile URLs.

Step 2 — Run signals

For each contact and each approved signal question:

saber signal --profile <linkedin-url> --question "<question>" --answer-type boolean

Step 3 — Review and prioritise

Present results to the user. Contacts where the signal fired positively should be sequenced first for outreach.

Key commands

saber signal --profile <linkedin-url> --question "<question>" [--answer-type] [--no-wait]
saber signal get <signalId>