Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add swan-gtm/gtm-skills --skill personalized-outbound-ab-testgit clone --depth 1 https://github.com/swan-gtm/gtm-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/swan-gtm/gtm-skills/personalized-outbound-ab-test)<a href="https://agentmods.dev/skills/swan-gtm/gtm-skills/personalized-outbound-ab-test"><img src="https://agentmods.dev/badge/skills/swan-gtm/gtm-skills/personalized-outbound-ab-test/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/swan-gtm/gtm-skills/personalized-outbound-ab-test"><img src="https://agentmods.dev/badge/skills/swan-gtm/gtm-skills/personalized-outbound-ab-test.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00070 | $0.00781 |
| Opus 5 | $0.00035 | $0.00391 |
| Sonnet 5 | $0.00014 | $0.00156 |
| Haiku 4.5 | $0.00007 | $0.00078 |
Grade A, and why
personalized-outbound-ab-test scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 13d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run this before you let an agent write outbound at scale. It produces a measured answer to "does personalization actually beat our generic message?" and a holdout that receives the winner.
The play
-
Split the list into three. Control (e.g. 100) gets the generic baseline. Personalized (e.g. 100) gets researched messages. Holdout (the rest) waits for the result, then gets the winner. Without a control you are not testing, you are just spending.
-
Build the do-not-contact list first. Exclude anyone already in an open thread, in the CRM as an active deal, or recently sequenced. Skipping this is how you send a "nice to meet you" to a customer. Do this before enrichment so you never pay to research someone you cannot contact.
-
Enrich only the Personalized arm. Pull current role, headline, employer, and recent activity per lead. The control arm needs nothing, so half your enrichment spend disappears.
-
Write one message per lead, then gate it. Every message must pass all six checks or it silently drops to the baseline:
- Account-safe — no links, no phone numbers, no mass-template feel.
- Factually grounded — every personal reference traces to retrieved data. Zero invented facts.
- Human — no AI tells, in the sender's own rhythm.
- Respectful — warm, no creepy over-familiarity.
- Rule-compliant — length cap, CTA placement, banned vocabulary, whatever the sender set.
- Recognizably theirs — the hook is something only that person would recognize as about them.
-
Re-verify programmatically. Do not trust the model's self-report. Recompute length, dash counts, CTA position, name casing, and banned phrases after generation. Anything that fails gets fixed or dropped to baseline.
-
Ship with a fallback. Set the generic message as the fallback for the personalized variable, so a missing field degrades to proven copy instead of sending
{PERSONAL_MSG}to a prospect. -
Read replies, not opens. Compare reply rate and positive-reply rate between the two arms. Send the winner to the holdout.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 13d ago First seen · 47 lines · 70 tokens per session scan A e2e337526976
personalized-outbound-ab-test is a skill published in the GitHub repository swan-gtm/gtm-skills (150 stars, last pushed 2d ago), licensed MIT. It adds 70 tokens to every session and 781 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
sdr-agent
Autonomous SDR that monitors pipeline, drafts outreach, qualifies inbound, and manages follow-ups.
afrexai-lead-hunter
Enterprise-grade B2B lead generation, enrichment, scoring, and outreach sequencing for AI agents. Find ideal prospects, enrich with verified data, score against your ICP, and generate personalized outreach — all autonomously.
cold-outreach
Use when writing a cold email or LinkedIn DM to a stranger and its cadence: first-touch copy under a word ceiling, 4-7 step bump sequences, per-inbox volume and warm-up limits, the compliant opt-out footer. NOT SPF/DKIM/DMARC setup (that is email-deliverability), NOT sourcing the list (that is lead-gen), NOT life…
distributor-outreach-email
Writes a sales outreach email to a film or content distributor introducing a completed project for acquisition consideration, including key commercial information an acquisitions team needs to evaluate the opportunity.
ai-cold-email
Generate personalized cold outreach emails using AI.
cold-email-personalizer
A tool that writes personalised cold emails, which are unsolicited messages sent to potential customers or partners, using information about each person.