From saber-skills
Design a multi-step outreach sequence (email + LinkedIn) with messages personalised using Saber signal results and timing tied to signal triggers.
How this skill is triggered — by the user, by Claude, or both
Slash command
/saber-skills:build-sequenceThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use this skill to design a full multi-step outreach sequence for a target persona, with each touchpoint personalised using Saber signal data and timing informed by signal triggers.
Use this skill to design a full multi-step outreach sequence for a target persona, with each touchpoint personalised using Saber signal data and timing informed by signal triggers.
Produce a complete sequence plan: N steps, each with a channel, timing, and a personalised message template that references specific signals.
Confirm from conversation context or ask:
If a signal-discovery run has already happened in this conversation, use those ICP and signal definitions directly.
Signal results personalise each step. Pull what's available before designing the sequence.
If the Saber CLI is available (saber --help works):
Check for existing subscription results covering the target accounts:
saber subscription list
saber subscription get <subscriptionId>
For contact-level context (if LinkedIn URLs are known):
saber signal --profile <linkedin-url> --question "Is this person posting about [relevant topic]?"
If no results exist yet, ask the user what signals they typically track and design the sequence around those signal hypotheses — placeholders can be filled in once signals run.
If the Saber CLI is not available: design the sequence with signal placeholder hooks (e.g. [IF hiring signal: ...]) that the user can personalise once they have data.
Ask the user their preferred sequence length and mix, or recommend based on the persona:
Recommended structure for cold outreach (B2B SaaS):
| Step | Day | Channel | Purpose |
|---|---|---|---|
| 1 | 0 | Signal-led first touch | |
| 2 | 3 | Connection request with context | |
| 3 | 7 | Value-add follow-up (resource or insight) | |
| 4 | 12 | Engage on their content (comment or message) | |
| 5 | 18 | Direct ask / break-up |
Adjust based on the persona (executives = shorter, fewer steps; SDR targets = more touches).
For each step, write a message template. Use the strongest available signal in Step 1, and distribute different angles across the sequence (don't repeat the same hook).
Step 1 — Signal-led first touch (email)
Use the highest-confidence signal as the opener. See write-outreach for format guidelines.
Step 2 — LinkedIn connection request Reference the email or a different signal angle. 300 chars max. No hard CTA.
Step 3 — Value-add follow-up (email) Don't push the product again. Share something useful: a relevant case study, a data point tied to their signals, a framework for the problem you solve. Soft CTA.
Step 4 — LinkedIn engagement If the contact has posted recently (use contact-level Saber signals or check LinkedIn), comment meaningfully on their post. If not, send a short message referencing the earlier email.
Step 5 — Direct ask / break-up (email) Be direct. "Is this a priority for you right now?" or "Should I stop reaching out?" — honest, low-friction. Often the highest-response step.
For each step, add a conditional variant based on key signals:
Step 1 — IF hiring signal positive:
"I saw [company] is scaling the sales team — most RevOps leaders at that stage are dealing with [problem]..."
Step 1 — IF funding signal positive:
"Congrats on the recent round — [company] is clearly in growth mode. Teams at this stage often prioritise [problem]..."
Step 1 — IF no strong signals:
[Fall back to persona-level pain without a specific event hook]
Output the complete sequence as a structured plan, each step clearly showing:
Offer to export as a format the user's sequencing tool (Outreach, Salesloft, Apollo, HubSpot sequences) can import.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub saberapp/saber-marketplace --plugin saber-skills