agentleFS
kyle-olmsteadkyle-olmstead/cs-claude-engine/skills/red-account-turnaround/SKILL.md

red-account-turnaround

Skill1 starsChanged 24 days agoApache-2.0View on GitHub

This skill should be used when the user says "[account] just went red", "run the turnaround on [account]", "our champion at [account] left", "[account] is at risk", "do a churn root-cause on [account]", "why are we losing [account]", "build a save plan for [account]", "red account review", or when a book review flags PB-RET-01 on an account. It runs the enterprise at-risk turnaround: a churn root-cause analysis (mechanism, timeline, blast radius, interrupting intervention) chained into a renewal defense plan (scenarios, stakeholder asks, dated actions, one approval-gated outreach draft), with every claim tied to a field in the Customer 360.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Follow ${CLAUDE_PLUGIN_ROOT}/skills/customer-360/references/playbook-protocol.md with these specifics.

    line 18
  2. Very specific:

    If actions_last_30d or history/actions.jsonl shows an open PB-RET-01 inside 14 days, emit a signal_update instead of a new action.

    line 30
  3. Very specific:

    If any NPS verbatim or call is newer than the last run and lacks a signal, run prompts/core/sentiment_analyzer.prompt on it first and merge …

    line 36
  4. Very specific:

    Before the RCA, read ${CLAUDE_PLUGIN_ROOT}/playbooks/enterprise_at_risk_turnaround.md §2 (severity tiers) and §7 (the save conversation) so …

    line 38

≈ “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

Red account turnaround (PB-RET-01)

Follow ${CLAUDE_PLUGIN_ROOT}/skills/customer-360/references/playbook-protocol.md with these specifics.

When it applies

Confirm at least one trigger holds on the Customer 360 (otherwise say which playbook fits better and offer it):

  • health.band == "red", or a drop of ≥ 15 points in 30 days
  • relationship.champion.status == "departed" with no replacement
  • a P1 open more than 5 days
  • renewal ≤ 120 days with health < 65
  • sentiment.ecri ≥ 60

If actions_last_30d or history/actions.jsonl shows an open PB-RET-01 inside 14 days, emit a signal_update instead of a new action.

Prompt chain

  1. ${CLAUDE_PLUGIN_ROOT}/prompts/retention/churn_risk_rca.prompt: produces the diagnosis. Build the dated timeline from the last 120 days before choosing a mechanism; name one interrupting intervention; include the disambiguating question when confidence < 0.75.
  2. ${CLAUDE_PLUGIN_ROOT}/prompts/retention/renewal_defense_playbook.prompt with rca_json = the RCA's diagnosis object: produces the plan. Evaluate all three scenarios; one kickoff/continuity draft only.
  3. If any NPS verbatim or call is newer than the last run and lacks a signal, run prompts/core/sentiment_analyzer.prompt on it first and merge the result into the document.

Before the RCA, read ${CLAUDE_PLUGIN_ROOT}/playbooks/enterprise_at_risk_turnaround.md §2 (severity tiers) and §7 (the save conversation) so the plan's internal escalation and first-touch owner match the tier: Critical → internal exec sponsor touches within 48h; High → CSM within 72h; Elevated → 5 business days.

Artifacts

  • PB-RET-01.churn_risk_rca.action.json: the RCA
  • PB-RET-01.renewal_defense_playbook.action.json: the plan
  • PB-RET-01.renewal_defense_playbook.comm_1.md: exec continuity or renewal kickoff draft (approval required)
  • An internal escalation brief: render from the two actions into PB-RET-01.escalation_brief.md when the tier is Critical or High. One page for leadership, written per ${CLAUDE_PLUGIN_ROOT}/skills/customer-360/references/writing-standard.md: the account in brief, what happened and why, what is at stake, what we need from whom by when, and the ask of leadership.

Proposed writes (offered, never automatic)

  • ~~crm: cse_risk_tier = red_<driver> on the account; task "Red account review: <account>" for the CSM and CS leader, due 2 days; cse_scenario on the renewal opportunity.
  • ~~chat: P0 alert to the red-accounts channel named in the stack profile (summary, confidence, items awaiting approval).
  • ~~calendar: a proposed hold for the exec continuity meeting inside the first-touch window.
  • ~~documents: save the RCA + plan as one document for the war room.

Closing the loop

Tell the user the exit criteria (band ≠ red for 30 days and renewal closed or > 180 days out; or churn confirmed) and that re-running this skill after the first customer conversation, with what was learned, refreshes the RCA against the open action rather than opening a second one.

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.