agentleFS
CSPulseCSPulse/customer-success-skills/skills/save-plan/SKILL.md

save-plan

Skill2 starsChanged 3 months agoMITView on GitHub

Use when the user says "save plan", "save this account", "can we still save them", "30 day rescue", "they are leaning toward leaving", "turnaround plan for". Produces a window verdict (launch a save or manage the exit), then a mechanism-matched 30 day plan by week, the executive sponsor, asks, a day 15 go or no-go and the contraction you would accept in advance. Not for judging whether they will renew (use renewal-risk), a customer who already gave notice (use offboarding) or a live angry escalation (use escalation).

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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
  1. Distinctive:

    If the user gave the account, ARR, renewal and notice dates, and what is going wrong, go straight to "Do this".

    line 13
  2. Distinctive:

    Use this skeleton. Keep it to one page.

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Save Plan

Turn an at-risk read into a 30 day plan with one owner, a checkpoint that can fail, and a landing you chose before the pressure started.

Start here

If the user gave the account, ARR, renewal and notice dates, and what is going wrong, go straight to "Do this". A renewal-risk read already in the conversation counts.

Otherwise ask for these in ONE message, then proceed with whatever comes back:

  1. Account, ARR, renewal date, notice date, and what the user has been told by whom
  2. Who decides and whether the user has met them
  3. What the user's company can offer: executive time, price or term room, services hours

Usage, emails and the contract are a bonus. If an "Account context" block exists in the conversation, use it.

Do this

  1. Decide if there is a window. Any one of these closes it: the decider has said no, the notice date has passed, a replacement is procured, the budget is reallocated. Closed means do not launch; manage the exit. Narrow means count the working days left and say so.
  2. Name the mechanism and match the play. Budget cut, failed rollout, sponsor lost or reorganized, competitor, price, service failure, product gap. Load references/plays.md. If the mechanism is unknown, week 1 is finding it.
  3. Set the people. One owner on the user's side. An executive sponsor at the level of their decider. The decider on their side, met in week 1, not only the champion.
  4. Put the fix and one proof first, the price last. One result in the decider's own measure, in writing, within two weeks. Commercial options come after, as trades.
  5. Pre-decide the landing. The contraction, term and add-back right the user would accept, and the floor below which they let it go. Get finance authority before the first call.
  6. Write it by week with the ask of the customer each week, the asks of the user's own company (name, action, date), and a day 15 checkpoint with a stated consequence if it fails.
  7. Log it where the next owner will read it: window, mechanism, owner, floor, checkpoint.

Output

Use this skeleton. Keep it to one page.

SAVE PLAN: [Account] | [ARR] | renews [date] | opt-out by [date]

Window: OPEN / NARROW / CLOSED (manage the exit)
Why: [the facts, said / seen / inferred]. Working days left: [n]
Mechanism: [3 sentences, cause to effect]. Play: [name]
Dollars: save at $[x]. Accept down to $[floor] if [terms]. Below that, let go.

People: Owner [name] | Our exec [name] | Their decider [name], met yes / no

| Week | Action | Owner | Ask of customer | Date |

Asks of my company: [name, action, by when]
Day 15 checkpoint ([date]): "If [observable] has not happened, [switch to fallback]."
Fallback (graceful landing): [contraction, terms, door-open step]
Log: [where the plan is recorded]

What I could not check
- [gap]

A filled example is in examples/sample.md. Match its length and tone.

Rules

  • No window, no plan. A save launched after a replacement is bought wastes a month and spends the goodwill a clean exit or a later win-back needs.
  • Match the play to the mechanism. A discount does not fix a failed rollout and training does not fix a cut budget. The wrong play feels busy and changes nothing.
  • Fix before price. Opening with a discount teaches the customer that threatening to leave works, and it answers none of their reasons.
  • Protect the relationship with the decider, not only the champion. A champion can be reorganized away overnight, and the signer is the one who reads the renewal.
  • One owner. A plan owned by three people has nobody holding the day 15 checkpoint.
  • The day 15 checkpoint must be able to fail. Half the runway is gone by then, and a test that cannot fail turns the plan into ritual while the notice date closes.
  • Choose the contraction in advance. Under pressure people concede whatever is asked. A floor decided beforehand makes a smaller account that renews a win, not a failure.
  • Never trade the unverified. Roadmap dates, price and security terms come from the user's company, not from this page. Flag them before they reach a draft.

Related (only if installed)

renewal-risk for the read this plan starts from. stakeholder-map when the decider is unclear. renewal-negotiation for the commercial rungs. offboarding when the window is closed. adoption-plan when the mechanism is low usage.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.