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 inbox-warm-up-rampgit 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/inbox-warm-up-ramp)<a href="https://agentmods.dev/skills/swan-gtm/gtm-skills/inbox-warm-up-ramp"><img src="https://agentmods.dev/badge/skills/swan-gtm/gtm-skills/inbox-warm-up-ramp/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/inbox-warm-up-ramp"><img src="https://agentmods.dev/badge/skills/swan-gtm/gtm-skills/inbox-warm-up-ramp.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 | $0.00079 | $0.00837 |
| Opus 5 | $0.00039 | $0.00418 |
| Sonnet 5 | $0.00016 | $0.00167 |
| Haiku 4.5 | $0.00008 | $0.00084 |
Grade A, and why
inbox-warm-up-ramp 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 9d 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
A new mailbox at full volume is a spam complaint with a countdown. This sets the ceiling, per inbox, and the conditions to raise it.
The play
- Ramp by the age of the inbox, not the age of the campaign. Week one at a handful of sends a day; roughly double each week; reach full volume at about four weeks. Every seat runs its own clock — a new person joining a mature team starts at week one.
- Get volume from more inboxes, never from one working harder. Two or three mailboxes on a domain, more domains when you need more volume. A single inbox pushed past its ceiling is the failure everyone repeats.
- Send inside the recipient's working hours, on working days. Mail arriving at 3am local time reads as automation to a person and to a filter.
- Keep the first touch plain. No links, no images, no tracking pixel, no attachment. Add them once a thread has a human in it.
- Watch the three numbers that end a ramp: hard bounces (stop and clean the list well before 3%), spam complaints (any is a lot), and the share of sends that reach zero engagement over a week. A ramp is paused by evidence, not by feel.
- Never warm up on the account you cannot lose. If cold volume must grow, put it on a separate sending domain and keep the primary one for real conversation.
- Tell the user why the cap is low, in their words. A silent limit reads as a broken tool, and the person will raise it blindly to make the silence stop. Say the number, say the schedule, and if you offer a skip, spell out what it risks before they choose it.
What good looks like
- The tell: reputation is per sending identity, and it is earned by behaviour that looks like a person's — modest volume, replies coming back, activity inside a working day. Volume without replies is the pattern filters are built to catch, which is why the ramp exists at all and why a sequence that gets answers can climb faster than one that does not.
- The mediocre version warms up by mailing seed accounts in a warming network and calls the inbox ready on day three. The metric moves; the reputation does not. Nothing detectable happens until real volume starts landing in spam, and by then the domain is spent.
- You know the output is good when raising the cap changes nothing except the count — same bounce rate, same complaint rate, same reply rate. If replies fall as volume rises, the ceiling was already correct.
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.
- 9d ago First seen · 69 lines · 79 tokens per session scan A beebea0c1e97
inbox-warm-up-ramp is a skill published in the GitHub repository swan-gtm/gtm-skills (153 stars, last pushed 2d ago), licensed MIT. It adds 79 tokens to every session and 837 once invoked, about $0.0004 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 2026-09-03.
Other skills, from other repositories
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.
outreach-strategist
Reviews and coaches your outreach with specific scoring, feedback, and rewrites.