welcome
Skill1 starsChanged 6 months agoMITView on GitHub
Interactive setup for the gtm-cold-email-pack. Interviews the sender, drafts config/sender-profile.md, optionally validates against a pasted LinkedIn or resume, and offers to symlink the pack's skills into ~/.claude/skills/. Run this once after cloning the pack. Trigger phrases — "welcome", "set up the cold email pack", "init my sender profile", "first time using this pack".
7 other projects use a version of this.
Adopted by
Most recently taken up by
- microsoft/amplifier-app-nanoclawtheir own version, first seen Oct 13, 2026
- TravisCao/cleanresearchtheir own version, first seen Oct 13, 2026
- Pearl1531/Archetype-talktheir own version, first seen Oct 13, 2026
- shorin-nikita/ai-teamtheir own version, first seen Oct 11, 2026
- tymrtn/tomorrowkit-for-mindstheir own version, first seen Aug 30, 2026
- Q00/ouroborostheir own version, first seen Jul 31, 2026
Other versions7 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 7
- Q00/ouroboros★ 6.3ksame name
- microsoft/amplifier-app-nanoclaw★ 5same name
- tymrtn/tomorrowkit-for-minds★ 4same name
- adraalabs/skills★ 2same name
- Pearl1531/Archetype-talk★ 2same name
- shorin-nikita/ai-team★ 1same name
- TravisCao/cleanresearch★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository8 other files it gives its agents.
- create-promptSkill
- create-skillSkill
- email-founderSkill
- email-growth-leaderSkill
- email-hiring-managerSkill
- humanizeSkill
- pick-frameSkill
- story-consistencySkill
The file
We split this skill into the rules it gives an agent
9 rules. 0 are common, 5 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 6 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 22
If it doesn't resolve, ask the user where they cloned the pack and cd there.
- Very specific: line 26
Read ./config/sender-profile.md.
- Distinctive: line 30
Ask: "Section-by-section interview (5–10 min), or paste a LinkedIn URL / resume / writing sample and I'll draft from that and only ask …
- Distinctive: line 32
If they paste source material, parse it for:
- Distinctive: line 53
If they paste a sample, analyze sentence length distribution, punctuation habits (em-dashes?
- Very specific: line 65
Write to ./config/sender-profile.md.
- Very specific: line 69
Ask: "Symlink the pack's skills into ~/.claude/skills/ so you can use them from any directory?
- Distinctive: line 71
If yes, show these commands (substitute the absolute path to the pack root) and wait for explicit confirmation before running:
- Distinctive: line 86
If the preview reads weak (generic bullet, voice off, hero vague), name the specific profile section that needs more detail and offer to …
“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Welcome — Quick-Start Setup
You are a world-class interviewer onboarding a new user to the gtm-cold-email-pack. Use AskUserQuestion for every interview question — one at a time, with concrete options where the answer space is finite. The output is only as good as the interview, so probe vague answers and reflect back what you heard.
The deliverable is config/sender-profile.md — the file every other skill in the pack reads. Get it right and the downstream skills (pick-frame, email-founder, email-growth-leader, email-hiring-manager, story-consistency, humanize) work on the first try.
Step 0: Locate the pack
The pack root contains config/sender-profile.example.md and skills/. Verify:
ls config/sender-profile.example.md skills/If it doesn't resolve, ask the user where they cloned the pack and cd there.
Step 1: Check whether a profile already exists
Read ./config/sender-profile.md. If it exists, ask: "Keep as-is, update specific sections, or start over?" Honor the answer; never overwrite without explicit confirmation.
Step 2: Offer the fast path
Ask: "Section-by-section interview (5–10 min), or paste a LinkedIn URL / resume / writing sample and I'll draft from that and only ask follow-ups?"
If they paste source material, parse it for:
- Most recent role + company →
Current status - Metric-led bullets (only with real numbers or named systems) →
Hero achievements - Tools / products / ecosystems mentioned →
Ecosystem connections - Writing voice patterns (sentence length, em-dashes, hedging, sign-offs) →
Voice rules
Then run Step 3 as a gap-filling pass — only ask where the source was thin.
Step 3: The interview
For every question below: ask once, listen, and if the answer is vague or generic, push back once with a specific follow-up before moving on. A "good answer" names a specific company, system, number, or product. A "bad answer" is generic ("drove growth", "led marketing", "exploring opportunities"). When you hit a bad answer, name what's missing and ask again.
3.1 Current status — "In one paragraph: what are you doing now, what did you just leave, what are you looking for?" Push for a target type of company (devtools, AI infra, B2B SaaS at a specific stage) — not just "growth roles."
3.2 Hero achievements (3–5 bullets) — "Format: [Outcome with a number] at [Company] — [one-line how]." If they give you a responsibility ("I owned X"), reframe: "What was the outcome?"
3.3 Ecosystem connections — "Products where you could legitimately open a cold email with 'I use this in prod.'" Push for the short, real list — not the exhaustive one. Ask: "Which of these could you talk about for 10 minutes without bluffing?"
3.4 Voice rules (In / Out) — Ask both: "Patterns to keep?" and "Patterns to forbid?" Then offer the calibration option: "Paste a recent email or LinkedIn post you wrote and liked. I'll extract patterns and ask you to confirm."
If they paste a sample, analyze sentence length distribution, punctuation habits (em-dashes? semicolons?), capitalization, sign-offs, and tone tells. Reflect back 3–5 patterns and ask them to confirm.
Always seed the Out list with these (ask the user to confirm, not just rubber-stamp): em-dashes, "I hope this finds you well," corporate hedging ("potentially," "leverage," "synergy"), three-option asks, research mirror-backs ("I saw you raised $X"). This is the highest-leverage section — a wrong voice setting means every email sounds off.
3.5 Dealbreakers — "What's a hard rule the email skills must never violate?" Probe: "Has anything you wrote ever landed wrong? What was the specific thing?" Always seed: "Never invent metrics or outcomes" and "Never claim domain insider status outside listed ecosystems."
Step 4: Draft and review
Compose the full profile in the structure of config/sender-profile.example.md. Show it to the user before writing the file and ask: "Anything to change before I save it?" Iterate until they say ship it. Drafts are cheap; surprises are expensive.
Step 5: Write the file
Write to ./config/sender-profile.md. Confirm the path. The file is gitignored — the profile stays local.
Step 6: Offer to install skills
Ask: "Symlink the pack's skills into ~/.claude/skills/ so you can use them from any directory? I'll show the commands first."
If yes, show these commands (substitute the absolute path to the pack root) and wait for explicit confirmation before running:
mkdir -p ~/.claude/skills
for skill in pick-frame email-founder email-growth-leader email-hiring-manager humanize story-consistency welcome; do
ln -sf "<absolute path to pack>/skills/$skill" ~/.claude/skills/$skill
doneln -sf overwrites existing symlinks — call that out. Offer cp -R as a fallback for users who prefer copying (caveat: they re-copy on updates).
Step 7: Show the customization
Prove the profile actually drives the downstream skills. Pick one Ecosystem connection and one Hero achievement from what the user just gave you, and inline a 4-line preview of /email-founder's opening + first bullet against a hypothetical target in that ecosystem. Highlight which voice rules were applied.
If the preview reads weak (generic bullet, voice off, hero vague), name the specific profile section that needs more detail and offer to re-run that part of the interview.
Step 8: Hand off the toolkit
Setup (re-run anytime)
/welcome — re-interview or update sections
Route (when unsure which drafter)
/pick-frame — picks the right drafter for a given target
Draft (pick one based on recipient)
/email-founder — founder/CEO with product or ecosystem tie
/email-growth-leader — Head of Growth / VP Growth (stage-matched builder)
/email-hiring-manager — JD-anchored fit email
Polish (run on any draft, in this order)
/story-consistency — fact-checks claims against your profile
/humanize — strips AI-tells, applies your voice rulesState the flow once: "/pick-frame → drafter → /story-consistency → /humanize → ship. Skip /pick-frame if you already know which drafter you want."
Close with: "If a draft feels off, the cause is almost always a vague section in sender-profile.md. Re-run /welcome → option 2 (update specific sections) anytime."
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

