write-the-follow-up
Skill5 starsChanged 4 months agoMITView on GitHub
Writes the next email after a call, a proposal, an event or a stalled thread, using what was actually said rather than a template. Use when someone has rewritten the same message four times, or when follow-ups all sound like checking in.
In this repository30 other files it gives its agents.
- AGENTS.mdAGENTS.md
- 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
The file
Show its rules and text
We split this skill into the rules it gives an agent
14 rules. About 0 are common, 14 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈14
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 12
Follow-ups fail for a boring reason: they contain no new information.
- Distinctive: line 35
Write it, then cut the first sentence.
- Distinctive: line 36
Check the ask is answerable in one line.
- Distinctive: line 41
No message without new information.
- Distinctive: line 43
Never invent urgency. A deadline that is not real gets discovered, and it costs the relationship more than the deal is worth.
- Distinctive: line 44
Do not apologise for following up.
- Distinctive: line 45
If the honest content of the message is "I want an update", write the closing message instead and let them reopen it.
- Distinctive: line 59
Always provide it. Roughly half the length.
- Distinctive: line 67
After a proposal sent nine days ago with no response, to a head of operations who writes in short, unpunctuated lines.
- Distinctive: line 75
Do not send anything. Produce the text.
- Distinctive: line 76
Do not fabricate case studies, customer names, results or deadlines to create urgency.
- Distinctive: line 77
Do not write a message whose only content is a request for a status update.
- Distinctive: line 78
Do not stack multiple asks and call it giving options.
- Distinctive: line 79
Do not claim a reply rate.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
Write the Follow-Up
Use this skill when
The next email has to go out and it has been rewritten three times already.
Follow-ups fail for a boring reason: they contain no new information. The recipient has nothing to react to, so reacting costs them effort with no return, so they do not. Everything below exists to give them something worth answering.
For a thread that has already gone silent for weeks, use ghosted-after-the-demo first to work out why, then come back here to write it.
Required input
- What happened: the call, the proposal, the event, the last message.
- The thread so far, if there is one.
- Who you are writing to and what they can actually decide.
Useful:
- what they said they cared about, in their words
- what was agreed as the next step and who owned it
- any deadline that is genuinely real
- how formal their own messages are
Analysis workflow
- Find the new information. Something must have changed since the last message: an answer to their question, a relevant example, a piece of thinking they have not seen. If nothing has changed, that is the finding, and the honest options are to create something worth sending or to wait.
- Establish what you are actually asking for. One thing. A meeting, an answer, an introduction, or permission to close the file.
- Match the length to the stage. After a first call, short. After a proposal, short. Nobody has ever complained that a follow-up was too brief.
- Mirror their register. If their emails are four words and no greeting, a three-paragraph message reads as a mismatch and slows the reply.
- Write it, then cut the first sentence. It is almost always throat-clearing, and the second sentence is usually the real opening.
- Check the ask is answerable in one line. If replying requires them to think, schedule or consult someone, the reply rate drops by an order of magnitude.
- Offer the exit. A message that makes it easy to say no gets more replies than one that only permits yes, and a clear no is worth more than another week of silence.
Decision rules
- No message without new information. This is the rule that fixes most follow-up problems on its own.
- One ask per message.
- Never invent urgency. A deadline that is not real gets discovered, and it costs the relationship more than the deal is worth.
- Do not apologise for following up. It signals that the message is an imposition, which invites the recipient to treat it as one.
- If the honest content of the message is "I want an update", write the closing message instead and let them reopen it.
Output format
The message
Ready to send, subject line included.
Why it is built this way
Three lines: what the new information is, what the single ask is, and what the exit is.
The shorter version
Always provide it. Roughly half the length. In practice it is the one that gets used.
If you get no reply
What the next message is and when, or a plain recommendation to stop.
Practical example
After a proposal sent nine days ago with no response, to a head of operations who writes in short, unpunctuated lines.
There is no new information yet, so the skill says so and creates some: a one-line answer to the integration question they raised on the call and which the proposal did not address directly.
The message is four lines. It leads with the integration answer, not with the proposal. The ask is whether that was the blocker. The exit is an offer to close the file if the timing has moved. The shorter version is two lines and drops the greeting to match their register.
Guardrails
- Do not send anything. Produce the text.
- Do not fabricate case studies, customer names, results or deadlines to create urgency.
- Do not write a message whose only content is a request for a status update.
- Do not stack multiple asks and call it giving options.
- Do not claim a reply rate. The message removes friction; it does not create intent.
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.

