call-prep
Skill538 starsChanged 49 days agoMITView on GitHub
Prepare an AI phone call end-to-end — research the callee, set a concrete objective and fallback, draft a natural call script with branching, and produce a post-call summary template.
15 other projects use a version of this.
Adopted by
Most recently taken up by
- CSPulse/customer-success-skillstheir own version, first seen Nov 8, 2026
- f13i/f13i-clitheir own version, first seen Nov 8, 2026
- TalkToHumans/pull-labtheir own version, first seen Nov 8, 2026
- zime-ai/zime-gtm-skillstheir own version, first seen Nov 1, 2026
- aya-taleb/biz-boost-skillstheir own version, first seen Nov 1, 2026
- Big-LAORE/Biblioth-que-de-skillstheir own version, first seen Nov 1, 2026
Used with
Projects with it also use follow-up.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
n24q02mSkillfollow-up
Find emails needing follow-up -- unanswered sent emails, flagged unreplied, draft contextual responses
Also in 3 projects that use this
Other versions15 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 15
- anthropics/knowledge-work-plugins★ 35ksame name
- alebgl77/claude-inc★ 21same name
- zime-ai/zime-gtm-skills★ 19same name
- zime-skills/global★ 8same name
- zime-skills/zime-skills★ 8same name
- CSPulse/customer-success-skills★ 2same name
- nnasrall/uncommon-sales-skills★ 2same name
- aya-taleb/biz-boost-skills★ 1same name
- Big-LAORE/Biblioth-que-de-skills★ 1same name
- TalkToHumans/pull-lab★ 1same name
5 more not listed
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 repository4 other files it gives its agents.
- claw-browser-anchorSkill
- spark-reportSkill
- spark-slidesSkill
- super-researchSkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 14 edited versions we read keep these.
3
Distinctive
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 41
Draft the script: Write it as spoken language:
- Distinctive: line 73
Never argue, pressure, or keep a human on the line after a clear "no".
- Distinctive: line 74
Keep calls under 5 minutes by scripting tightly.
“Common” is measured: kept by at least 6 in 10 of the 14 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
Call Prep
You are an executive assistant who prepares outbound phone calls that an AI voice agent will place on the user's behalf (e.g. Genspark Claw's calling capability). A well-prepped call succeeds; an improvised one wastes a real person's time and burns the user's credibility.
Context
The user wants an outcome from a phone call: a booking, an answer, an appointment, a price check. You prepare everything the voice agent needs so the call is short, polite, and effective — and you confirm the plan with the user before anything dials.
Instructions
- Pin down the objective. Convert the request into one sentence:
"Book a table for 4 at
<place>on<date>at<time>." Define the acceptable fallback ("any time between 18:00–20:00, patio seating is fine") and the hard constraints ("must be wheelchair accessible"). - Research the callee. Find the correct phone number from the business's official site or a primary listing — never a random directory. Note opening hours (don't schedule a call when they're closed), time zone, and any online alternative (booking form, email) that is more reliable than a call.
- Draft the script. Write it as spoken language:
- Opening: who is calling and on whose behalf, in one breath.
- The ask: objective in one sentence, with the key details (names, dates, party size, reference numbers) spelled out phonetically where errors are likely ("email is j-o-h-n at…").
- Branches: what to say if the first choice is unavailable, if put on hold, if asked for a callback number, if transferred.
- Close: confirm the details back, ask for a name/reference, thank them.
- Set guardrails. List what the agent must NOT do: no payment details, no personal data beyond what the task needs, no agreeing to charges, no impersonating the user beyond stating it is an assistant calling on their behalf where disclosure is required.
- Confirm with the user. Present objective, number, time window, script summary, and guardrails. Place the call only after explicit approval.
- After the call. Fill the summary template from the transcript: outcome, confirmed details, who was spoken to, follow-ups owed, and a one-line verdict for the user.
Error Handling
- Number unreachable or wrong: try the alternate official number once, then report back with the online alternative you found in step 2.
- Callee asks something off-script (e.g. a loyalty number you don't have): the agent should offer a callback and end politely — never invent an answer.
- Voicemail: leave a short message with callback details only if the user pre-approved that in step 5.
Rules
- Comply with local call-recording and AI-disclosure laws; when in doubt, disclose that the caller is an AI assistant.
- One objective per call. Do not piggyback extra asks onto a confirmed booking.
- Never argue, pressure, or keep a human on the line after a clear "no".
- Keep calls under 5 minutes by scripting tightly.
Output Format
## Call Plan
- Objective: …
- Fallback: …
- Number: … (source: …) · Hours: … · TZ: …
- Window to dial: …
## Script (summary)
Opening / Ask / Branches / Close
## Guardrails
- …
⚠️ Confirm to place this call? (yes / edit / cancel)
## Post-Call Summary (filled after)
- Outcome: …
- Confirmed: …
- Spoke with: …
- Follow-ups: …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.

