retain
Skill1 starsChanged 18 days agoMITView on GitHub
Post-delivery retention loop: scheduled check-ins, monthly value notes, quarterly business reviews, delight-peak review asks, referral and repurchase offers, churn-watch signals, retainer SLAs, proactive 60-day contract renewals, and client health telemetry — routed through ten modes. Use when asked to follow up after delivery, prove ongoing value, run a QBR, ask for a review, win a referral, manage client retainer SLAs, audit account health scores, or execute a 60-day retainer contract renewal with 3-tier expansion. Not for churn strategy itself (growth retention mode) or status reporting (client-comms).
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
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 38
Every invocation resolves to exactly one mode.
- Distinctive: line 53
Only the resolved mode's reference is loaded — the rest stay on disk, saving tokens on every run.
- Distinctive: line 113
Check-in with no reference to what was delivered — generic "just checking in" trains clients to ignore you.
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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
🔁 retain — Post-Delivery Retention Loop
One head skill for everything after handoff. Delivery wins the project; the loop wins the next one. Cadence runs delivery+7d → monthly note → QBR w/ transcript → review-ask at delight peak → referral/repurchase offer. Churn signals (usage drop, silence >14d) break the cadence and route to growth churn mode.
Modes
Every invocation resolves to exactly one mode. Match the request, then load only the matched reference:
| Mode | Trigger phrases | Behavior | Reference |
| check-in | "check in", "follow up after delivery", "delivery+7d", "how is it going" | Scheduled post-delivery check-in (delivery+7d, then monthly heartbeat) | references/check-in.md |
| value-note | "value note", "prove value", "monthly update", "what did we deliver" | Monthly value note tying shipped work to client outcomes | references/value-note.md |
| qbr | "qbr", "quarterly review", "quarterly business review" | Quarterly business review run from a transcript, decisions logged | references/qbr.md |
| review-ask | "ask for review", "testimonial", "review request", "google review" | Review/testimonial ask timed at the delight peak — this skill owns the review pipe | references/review-ask.md |
| referral-rebuy | "referral", "refer a friend", "repurchase", "repeat purchase", "renewal offer" | Referral or repurchase/renewal offer after value is proven | references/referral-rebuy.md |
| churn-watch | "gone quiet", "churn risk", "usage drop", "silence", "at-risk client" | Churn-signal triage (usage drop, silence >14d) → growth churn mode | references/churn-watch.md |
| sla | "sla", "warranty", "bug or feature", "retainer hours", "rollover hours", "response time" | Retainer SLAs, 30-day post-launch warranty governance, bug vs enhancement triage, and capacity burn-down tracking | references/sla.md |
| renewal | "retainer renewal", "contract renewal", "60-day renewal", "scope creep retro", "expand retainer", "3-tier renewal" | Proactive 60-day retainer renewal cadence, cumulative ROI receipts ledger, scope creep retro, and 3-tier expansion proposals | references/renewal.md |
| health | "client health", "account health", "churn radar", "health score", "stop the clock", "account telemetry" | Tri-vector health scoring (Product 40%, Sentiment 30%, Financial 30%), Churn Early Warning Radar, and 48h Stop-the-Clock interventions | references/health.md |
| audit | "churn audit", "retention audit", "audit portfolio risk" | Five-question churn audit + leading-indicator rubric across the book; predictability and instrumentation gaps. | references/audit.md |
Only the resolved mode's reference is loaded — the rest stay on disk, saving tokens on every run.
When to Use
- Following up after delivery (delivery+7d check-in, monthly heartbeat).
- Proving ongoing value with a monthly value note.
- Running a quarterly business review from a call transcript.
- Asking for a review or testimonial at the delight peak.
- Making a referral or repurchase/renewal offer.
- Spotting churn risk from usage drops or client silence.
Anti-Triggers
- Churn strategy and save plays →
growthretention mode (churn-watch routes there). - Status reporting on active work →
client-comms. - Delivery retrospectives looking backward →
opsretro. - Measuring retention metrics →
analytics.
Quick Reference
Retention cadence (default order)
| Step | Mode | Timing |
| 1. Prove it landed | check-in | delivery+7d |
| 2. Prove it keeps paying | value-note | monthly |
| 3. Prove it strategically | qbr | quarterly, transcript in |
| 4. Capture the delight | review-ask | delight peak (delivery +7d, QBR close) |
| 5. Compound it | referral-rebuy | after value proven, never before |
Interrupt: churn signals at any step → churn-watch → growth churn mode.
Tooling
Mechanisms are tool-independent. Default stack where a stack was chosen: check-ins via whatsapp-business + gmail; QBR transcripts via fathom; searches modern-first (rg, fd) with legacy fallback.
Suite contracts
- Churn strategy and save/win-back plays →
growthretention mode. - QBR transcript capture → fathom pipe; decisions →
ops+evidence-ledger. - Retention metrics and dashboards →
analytics. - Active-work status →
client-comms.
Procedure
- Intake. Establish delivery date, client, and last contact before running anything — retention without a timeline spams.
- Resolve the mode. Match the request against the mode table; exactly one mode. Ambiguous → ask one question, then proceed.
- Execute the mode playbook. Load
references/<mode>.mdand follow its Intake → Deliverable → Procedure → Quality gate in order. - Gate and deliver. Pass this file's Verification checklist plus the mode's quality gate, then deliver the artifact where the client expects it.
Pitfalls
- Check-in with no reference to what was delivered — generic "just checking in" trains clients to ignore you.
- Value note listing activity instead of outcomes — hours logged is not value proven.
- QBR with no transcript — decisions evaporate; every QBR is transcribed or it didn't happen.
- Review ask at a low point — never ask during an incident or an unpaid invoice.
- Referral offer before value is proven — asking for social capital you haven't earned.
- Sitting on churn signals — silence >14d or a usage drop routes to
growthimmediately, not next quarter. - Hand-edited claims with no source — every value claim traces to shipped work or a transcript.
Verification
- Delivery date and last contact established before outreach.
- Exactly one mode resolved and its reference followed.
- Value claims trace to shipped work, metrics, or a transcript.
- QBR decisions logged to
ops/evidence-ledger. - Review ask timed at a delight peak, never at a low point.
- Referral/rebuy offered only after value proven.
- Churn signals routed to
growthchurn mode, not sat on.
In this repository49 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .github/copilot-instructions.mdCopilot instructions
- llms.txtllms.txt
- accountsSkill
- analyticsSkill
- automationSkill
- brandSkill
- client-commsSkill
- contentSkill
- crmSkill
- databaseSkill
- designSkill
- devopsSkill
- growthSkill
- gtmSkill
- incident-responseSkill
- mobileSkill
- opsSkill
- paidadsSkill
- qa-launchSkill
- researchSkill
- sales-enablementSkill
- seoSkill
- smmSkill
- telegramSkill
- webdevSkill
- context-anchorSkill
- coupling-routerSkill
- evidence-ledgerSkill
- relaySkill
- secretarySkill
- gitSkill
- new-projectSkill
- updateagentsSkill
- updatedocsSkill
- animateSkill
- designscopeSkill
- refactorSkill
- auditSkill
- code-reviewSkill
- dead-letterSkill
- gauntlet-loopSkill
- humanizeSkill
- muse-securitySkill
- puaSkill
- clean-system-cacheSkill
- coachSkill
- periodic-retreatSkill
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.