weekly-content-ideas-from-calls
Skill1 starsChanged 17 days agoMITView on GitHub
Use this skill at the end of the week when next week's content is still a blank page and the only raw material worth using is what clients and prospects actually said on calls. Produces 8 to 12 ranked content ideas, each with the verbatim quote behind it, a hook, and two formats to write it in, plus the questions to answer before drafting. Also use for "Friday content pull", "what should I post next week", "mine my calls for content", "turn client calls into LinkedIn posts".
In this repository67 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- morning-paper-outreachSkill
- client-activation-checkinSkill
- client-churn-autopsySkill
- client-win-backSkill
- late-payment-chaseSkill
- network-path-mappingSkill
- rate-increase-planSkill
- sales-call-debriefSkill
- weekly-funnel-reviewSkill
- buyer-readSkill
- punch-upSkill
- call-topic-miningSkill
- ceo-exec-assistantSkill
- og-image-from-templateSkill
- weekly-site-and-ad-reportSkill
- icp-reviewSkill
- client-performance-reviewSkill
- cold-email-sequenceSkill
- email-content-reviewerSkill
- scoring-analysis-and-distributionSkill
- delivery-audit-discovery-callSkill
- directionSkill
- rundownSkill
- source-checkSkill
- work-reviewSkill
- demandbird-exact-time-linkedin-postSkill
- kondo-owed-reply-sweepSkill
- linkedin-export-contact-importSkill
- mail-messages-contact-touchesSkill
- riverside-short-social-scheduleSkill
- youtube-transcript-panel-pullSkill
- email-voice-setupSkill
- upsell-opportunity-briefSkill
- ad-creative-content-strategySkill
- klaviyo-account-auditSkill
- facilitator-notesSkill
- local-prospectingSkill
- weekly-linkedin-postsSkill
- linkedin-watchlist-briefSkill
- pipeline-by-scriptSkill
- pre-session-flash-cardSkill
- write-in-the-owners-voiceSkill
- reddit-gtm-sprintSkill
- process-captureSkill
- sales-call-market-analyzerSkill
- linkedin-shoutout-postSkill
- plain-language-copy-checkSkill
- revenue-goal-funnel-mathSkill
- case-study-buildSkill
- cash-runway-forecastSkill
- client-profitability-readSkill
- client-proposalSkill
- content-briefSkill
- quarter-goal-filterSkill
- win-to-anglesSkill
- activity-incentive-sprintSkill
- candidate-interview-coachSkill
- client-list-stack-pitchSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
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 42
MUST keep every idea anchored to a verbatim quote from an actual call.
- Distinctive: line 43
MUST strip client and company names before anything is published, keeping role, size and situation.
- Distinctive: line 44
MUST rank on comments per 1,000 impressions, never on likes or follower count.
- Distinctive: line 45
MUST give each idea two formats.
- Distinctive: line 46
MUST run on the same day every week.
- Distinctive: line 47
NEVER publish a quote that identifies the speaker without their explicit yes.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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
Applies on a fixed day at the end of the week, before next week's content is planned. Produces a ranked idea list built only from things real buyers said out loud.
The play
- Pull every call transcript from the last seven days: sales calls, client calls, partner calls, internal strategy calls where a client problem was discussed. Nothing older. Stale weeks produce generic ideas.
- Read for moments, not topics. A moment is a sentence a buyer said that carries tension, a number, a contradiction, or a phrase in their own words. The types that keep earning, with examples, are in references/moment-types.md. Expect 15 to 30 candidate moments from a normal week and far fewer from a week of status calls.
- Pull each one as a verbatim quote with who said it (role, not name) and what prompted it. Paraphrase at this stage and the idea goes flat by Tuesday.
- Score every candidate against the ranking model in references/ranking-model.md. The metric is comments per 1,000 impressions on past posts, not likes and not raw impressions: it is the only one that tracks whether the audience had something to say back. Build the baseline once from the last 90 days of posts and reuse it.
- Cut to 8 to 12 ideas. Fewer than 8 means the week's calls were too thin, so widen to 14 days once and say so. More than 12 and nothing gets written well.
- For each surviving idea write: the quote, a hook line, and two formats it could be written in. Two formats, not one, because the same moment lands differently as a story than as a list, and the choice is made at drafting time.
- Flag the top 3. These are the ones that get written if the week collapses. Branch by what the week produced:
- A moment repeated by two or more different buyers goes straight to the top 3, whatever it scores.
- A single client's very specific situation scores well but is not a top 3 pick unless the pattern shows up elsewhere.
- A moment that contradicts something already published is always a top 3 pick. Those are the posts that get comments.
- Close with 6 to 8 voice questions, drawn from references/voice-questions.md and pointed at the top 3. They ask for what the transcript can't hold: what the writer was thinking but didn't say, the scene, the number nobody mentioned. The writer answers them out loud, often on a walk, and the draft is written from that recording, so the post comes out in their own words instead of as a summary of the call.
What good looks like
- First thing I check that most people skip: whether the quote would survive being read aloud to the person who said it. If it would embarrass them, it is not usable, however good the hook is.
- The mediocre version: a list of topics lifted from call summaries. "Clients are struggling with attribution." True, generic, already posted by everyone. The quote is what makes it specific, and the mediocre version drops it.
- I know the output is right when a writer can open the list cold on Monday and start typing without listening to a single call.
- Second test: at least two of the ideas make the writer slightly uncomfortable. Comfortable ideas get no comments.
- Red flag: the top 3 are all from the same call. That is one conversation, not a week of signal.
Rules
- MUST keep every idea anchored to a verbatim quote from an actual call.
- MUST strip client and company names before anything is published, keeping role, size and situation.
- MUST rank on comments per 1,000 impressions, never on likes or follower count.
- MUST give each idea two formats.
- MUST run on the same day every week. The cadence is what makes the backlog compound.
- NEVER publish a quote that identifies the speaker without their explicit yes.
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.

