Purpose
Turn messy case notes, ticket threads, or raw activity logs into polished, consistently structured status updates — one version for internal stakeholders and one for customers — so case managers can push out clear, professional updates fast without rewriting from scratch for every case.
When to Use
Invoke this skill when the user asks to:
- Write a status update for one or more open cases
- Summarize case notes into a clean progress update
- Draft a customer-facing update on a case, ticket, or claim
- Draft an internal update for a supervisor, team lead, or cross-functional partner
- Clear a backlog of pending case updates quickly and consistently
- Standardize how their team writes and formats case updates
Inputs Required
Before starting, collect:
- Raw case material: The notes, ticket thread, activity log, or data dump to convert
- Case identifier: Case/ticket number and (if relevant) the customer or account name
- Audience: Internal stakeholder, customer-facing, or both versions
- Case type: What kind of case this is (support ticket, claim, escalation, service request, investigation, etc.)
- Current status: What has just happened or what stage the case is at, if not obvious from the notes
- Known next step or owner: If already decided — otherwise the skill infers from the notes
- Tone preference: Neutral-professional / warm-reassuring / firm-direct (defaults to neutral-professional)
If audience is not specified, produce both an internal and a customer-facing version. If case type is unclear, infer it from the notes and state the assumption.
Instructions
Step 1: Parse and Classify the Raw Material
Read the raw case material and extract the essential facts:
- What has happened since the case opened or was last updated (actions taken, findings, communications)
- What is currently blocking or pending (waiting on customer, waiting on internal team, waiting on vendor, under review)
- What is decided vs. still open
- Strip filler, duplicate log entries, and internal shorthand that won't read cleanly.
Assign a single status classification from this standard set:
- New / Triaging — received, not yet actioned
- In Progress — active work underway
- Pending – Customer — awaiting customer response or action
- Pending – Internal/Third Party — awaiting another team or vendor
- Escalated — elevated for priority or specialist handling
- Resolved — closed with outcome
- On Hold — intentionally paused
Pick exactly one. If the notes suggest more than one, choose the one that reflects the immediate blocker.
Step 2: Build the Core Update Facts
Assemble the reusable facts both versions will draw from:
- One-line summary: A single sentence capturing where the case stands right now
- What we did: 2–4 bullet points of concrete actions or progress, in plain past tense
- Current status: The classification from Step 1 plus a one-line reason
- Next step: The single most important next action, with an owner (us or the customer) named
- Expected resolution / next update date: A specific date or timeframe — never "soon" or "shortly"
If any of these cannot be determined from the notes, flag it explicitly rather than inventing it.
Step 3: Generate the Internal Stakeholder Version
Write for a supervisor, team lead, or cross-functional partner who needs signal fast:
- Lead with the status classification and one-line summary
- Include specifics the customer wouldn't see: root cause notes, internal blockers, risk flags, effort/time spent, escalation triggers
- Be direct — no softening, no reassurance language
- Call out anything requiring a decision or help from the reader
Step 4: Generate the Customer-Facing Version
Write for the customer or external stakeholder:
- Open with a clear statement of where things stand, in accessible language
- Translate internal jargon into plain terms; remove internal names, tools, and blame
- Confirm what has been done and what happens next, with a date
- Set expectations honestly — if delayed, say so and why, briefly
- Close with a clear point of contact or next-update commitment
- Apply the tone preference; default to neutral-professional and reassuring without over-apologizing
Step 5: Handle Batches (If Multiple Cases)
When given multiple cases:
- Process each one through Steps 1–4
- Keep formatting identical across all cases so the batch reads as one consistent set
- Present them in a numbered sequence, each with its case identifier as a header
Output Format
- Structure: Per case — a short header (case ID + status classification), then the internal version and/or customer-facing version as labeled blocks
- Length: 120–200 words per version per case; scale linearly for batches
- Format: Structured update — one-line summary, short bulleted "what we did," and a clear next-step line; prose for the customer-facing body
- Sections: Case header → Status classification → Internal version → Customer-facing version → Next update date
- Tone: Internal version is direct and signal-dense; customer-facing version is clear, calm, and professional — never vague, never over-apologetic
Quality Checklist
Before delivering output, verify:
- [ ] Each case has exactly one status classification from the standard set
- [ ] Every update names a specific next step with an owner (us or customer)
- [ ] Every update includes a concrete date or timeframe — no "soon," "shortly," or "ASAP"
- [ ] The customer-facing version contains no internal names, tools, jargon, or blame
- [ ] Any fact not present in the source notes is flagged, not invented
- [ ] Batch updates use identical structure and formatting across every case
Notes
- The most common failure in case updates is a vague next step — always force an owner and a date, even if the date is "next update by [specific day]."
- If the raw notes contain conflicting information, surface the conflict in the internal version rather than smoothing it over in the customer version.
- For sensitive or escalated cases, keep the customer-facing version shorter and more measured — less detail reduces the risk of committing to something the team can't deliver.
- This skill standardizes format; it does not make case decisions. When the notes don't support a next step, say what's needed to determine one.
How to use this skill
This file is a Claude Code skill in Anthropic's standard format — markdown with YAML frontmatter at the top.
Installing in Claude Code:
~/.claude/skills/<folder-name>/SKILL.md— auto-loads in every session anywhere on your machine (recommended for skills you'll use across multiple projects)<your-project>/.claude/skills/<folder-name>/SKILL.md— auto-loads only when working in that project (recommended for project-specific skills)
Restart Claude Code or open a new session to register.
Using in Claude.ai, ChatGPT, Gemini, or other AI tools:
- Drag this file into a new chat, or copy and paste its contents
- Send your task or request as your next message — the AI will apply the skill where it fits
For Claude Code skill details: https://docs.anthropic.com/en/docs/claude-code/skills