b2b-lead-generation-claude-skills
AGENTS.md5 starsChanged 4 months agoMITView on GitHub
A pack of 30 Claude Skills for B2B lead generation. Each skill is a self-contained SKILL.md in its own folder, following the Agent Skills standard, so the pack runs in any Agent Skills tool, not only Claude Code. - Evidence first. Every skill works from what you supply: a CRM export, a transcript, an email thread, a DNS record, a company website. Where evidence is missing the skill marks the finding as an assumption rather than filling the gap with…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 32,671 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository30 other files it gives its agents.
- ad-to-landing-promise-matchSkill
- b2b-audience-signal-auditSkill
- cold-outbound-sequence-reviewSkill
- competitor-lead-angle-reviewSkill
- crm-lead-source-quality-auditSkill
- discovery-call-gap-analysisSkill
- disqualification-reason-minerSkill
- email-nurture-sequence-reviewSkill
- form-friction-finderSkill
- ghosted-after-the-demoSkill
- google-ads-lead-quality-auditSkill
- icp-reality-checkSkill
- landing-page-qualification-reviewSkill
- lead-magnet-offer-fitSkill
- lead-scoring-sanity-checkSkill
- lead-to-opportunity-drop-off-reviewSkill
- linkedin-ads-lead-quality-auditSkill
- meeting-prep-five-minutesSkill
- meta-ads-lead-quality-auditSkill
- objection-cheat-sheetSkill
- pipeline-hygiene-auditSkill
- proposal-win-loss-reviewSkill
- sales-follow-up-speed-auditSkill
- scale-readiness-check-lead-genSkill
- sdr-handoff-brief-generatorSkill
- spam-folder-checkSkill
- weekly-b2b-lead-gen-readoutSkill
- what-to-say-to-this-companySkill
- won-deal-to-case-studySkill
- write-the-follow-upSkill
The file
We split this AGENTS.md into the rules it gives an agent
7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Evidence first: Every skill works from what you supply: a CRM export, a transcript, an email thread, a DNS record, a company website.
- Distinctive: line 13
Nothing goes out: No skill sends an email, edits a CRM record, changes a sequence or touches DNS.
- Distinctive: line 16
No invented proof: No fabricated customer quotes, trigger events, funding rounds, case study numbers or reply-rate promises.
- Distinctive: line 76
Do not run a skill without its required input.
- Distinctive: line 77
Do not present an inference about a prospect's internal situation as fact.
- Distinctive: line 78
Do not treat one ghosted deal, one bad call or one lost deal as a pattern.
- Distinctive: line 79
Do not send, publish or apply anything a skill produced without a human reading it first.
≈ “Common” is an estimate: at least 1 in 5 of the 41,581 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AGENTS.md: B2B lead generation skills
A pack of 30 Claude Skills for B2B lead generation. Each skill is a
self-contained SKILL.md in its own folder, following the Agent Skills
standard, so the pack runs in any Agent Skills tool, not only Claude Code.
How these skills behave
- Evidence first. Every skill works from what you supply: a CRM export, a transcript, an email thread, a DNS record, a company website. Where evidence is missing the skill marks the finding as an assumption rather than filling the gap with a plausible number.
- Nothing goes out. No skill sends an email, edits a CRM record, changes a sequence or touches DNS. They produce text and recommendations that a human sends or applies.
- No invented proof. No fabricated customer quotes, trigger events, funding rounds, case study numbers or reply-rate promises. This matters more in B2B than elsewhere, because the buyer usually knows the truth about their own company and notices immediately.
How the pack splits
Skills 1 to 20 are the inbound and paid machine: targeting, pages, forms, scoring, source quality, handoff. They assume leads arrive and ask whether the right ones are arriving.
Skills 21 to 30 are the outbound week and the sales side. They assume nothing arrives and ask what you send, what you say on the call, and what you do with the silence afterwards.
The two halves answer different questions and are rarely useful at the same moment. A team with plenty of leads and a quality problem lives in the first twenty. A team with an empty pipeline lives in the second ten.
Order that matters
spam-folder-checkruns beforecold-outbound-sequence-review, always. A delivery problem and a copy problem produce the same symptom, and rewriting copy fixes only one of them.ghosted-after-the-demohandles one silent deal. When the same silence appears across many deals at the same stage, it stops being a message problem anddiscovery-call-gap-analysistakes over.pipeline-hygiene-auditcomes before any forecast conversation, not after one goes wrong.objection-cheat-sheetandwon-deal-to-case-studyboth read historical deals, so they are cheap to run together on the same export.
Where judgement shifts
- Deal size. Under a few thousand, most of the second ten are too heavy;
the volume game is the first twenty. Above five figures, per-account work like
what-to-say-to-this-companyandmeeting-prep-five-minutespays for itself on a single meeting. - Sales cycle length. A long cycle makes
pipeline-hygiene-auditessential and makes stale-deal thresholds much more forgiving. Set--cycle-daysrather than accepting a default. - Who does the selling. A founder selling directly needs the prep and follow-up skills. A team with SDRs needs the sequence, transcript and pipeline skills, because the failure is consistency rather than capacity.
- Regulated markets. Cold outreach rules differ by jurisdiction. No skill in this pack gives legal advice, and several say so explicitly.
Scripts
Two skills carry deterministic helpers, because their work is counting and dates
rather than judgement: pipeline_hygiene.py and email_auth_check.py.
selftest.py locks what they guarantee, including the defect where a future
close date was allowed to anchor the reference date and marked healthy deals as
overdue. Run it before and after changing anything in scripts/.
The other eight skills read text and produce text. A script there would add ceremony, not rigour.
What not to do
- Do not run a skill without its required input. Skip it and say what is missing.
- Do not present an inference about a prospect's internal situation as fact.
- Do not treat one ghosted deal, one bad call or one lost deal as a pattern.
- Do not send, publish or apply anything a skill produced without a human reading it first.
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.

