
customer-workspace
Set up and operate customer workspaces: route customer updates, record meetings, track questions and actions, manage workstreams, and prepare handoffs. Use when creating or maintaining a customer workspace, processing meeting notes or updates, reviewing next actions, or preparing a handoff. Triggers: customer workspace, 顧客ワークスペース, inbox 追加, 議事録管理, 案件更新, 宿題管理, 引き継ぎ.
- Installs
- 57
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/aktsmm/agent-skills --skill customer-workspace- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads customer-workspace automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
Customer Workspace
Set up or operate customer workspaces without confusing setup artifacts with live records.
When to Use
- Creating a per-customer workspace.
- Operating an existing workspace: inbox updates, meeting notes, questions, actions, workstreams, or handoffs.
- Adding customer-information routing or meeting-note workflows.
- Reviewing the current owner, next action, and source records before a handoff.
Choose a Mode First
Inspect the workspace before choosing an action. Treat it as existing when it contains any managed record such as .github/copilot-instructions.md, .github/prompts/, workspace-summary.md, _customer/profile.md, _inbox/, or workstreams/.
An aggregate operations workspace may contain a lightweight Customers/{id}/ profile, task list, inbox, and meetings. Do not initialize a full customer workspace over those records; ask for a separate target if dedicated customer management is needed.
| Mode | Use when | First action |
|---|---|---|
| Setup | No managed structure exists | Create the scaffold and capture routing facts. |
| Operate | A managed workspace receives new information | Update the canonical record for the input. |
| Review / Handoff | Status, ownership, or continuity is requested | Read current indexes first and report sources, owner, and next action. |
- Do not rerun the initializer or use
-Forcein an existing workspace unless overwriting generated assets is explicitly requested. - Existing records are authoritative. Add only explicitly requested missing assets; never replace live notes, tasks, or workstream records with templates.
Setup a New Workspace
# Basic
.\scripts\Initialize-CustomerWorkspace.ps1 -CustomerName "Contoso Inc"
# Full options
.\scripts\Initialize-CustomerWorkspace.ps1 `
-CustomerName "Contoso Inc" `
-ContractType "Ongoing support" `
-ContractPeriod "2025/04 - 2028/03" `
-KeyContacts "John Doe (Infra Lead)"
If PowerShell is unavailable, manually create the same folders and copy only the required prompt and template files from assets/.
Use setup only when the managed structure is absent. The initializer stops by default when managed assets exist; use -Force only for an explicitly approved replacement of generated prompts and templates.
Setup Intake
Capture only routing facts: workspace scope, sharing boundary, own-team aliases, key roles, meeting cadence, and primary inputs. Leave detailed technical inventory to later inbox and meeting records.
Canonical Records
| Need | Canonical record |
|---|---|
| Navigation and handoff | README.md, workspace-summary.md |
| Raw updates and open questions | _inbox/, _questions/ |
| Customer context | _customer/profile.md |
| Meetings and follow-up | meeting-notes/, next-actions/ |
| Ongoing scope and detailed history | workstreams/, pj_{topic}/ |
| Deliverables and reusable learning | research-reports/, _knowledge/ |
Initialize-CustomerWorkspace.ps1 creates the managed scaffold and its templates. Create workstream folders, meeting notes, next actions, project threads, and material folders only when the operating workflow requires them.
Leave cross-customer activity reports and global task indexes in the owning operations workspace; link to them when needed instead of copying them into a customer record.
Research Reports
Use research-reports/ for generated Markdown deliverables; keep the workspace root for entry files and controls. Detailed placement and sanitization rules are in Knowledge Ledger Rules.
Knowledge Ledger
Use _knowledge/ only for compact, reusable learnings when the user explicitly requests extraction or generalization. Apply Knowledge Ledger Rules before writing.
Optional Material Folders
Separate immutable originals, internal edits and shareable copies; use _received/, _working/ and _provided/ unless the user specifies another destination. Import accessible originals before relying on summaries. Renaming is not anonymization; restricted source content must stay out of recipient files, including hidden slides and notes. Apply Customer Material Lifecycle before moving or renaming files.
Operate an Existing Workspace
| Input or request | Action |
|---|---|
| Short chat, email, or unstructured update | Preserve it in _inbox/ and classify it. |
| Meeting memo or Teams AI record | Create or update one meeting note and extract questions/actions in the same operation. |
| Clear update for one confirmed workstream | Update its README with the source link and state change. |
| New, ambiguous, or multi-workstream input | Record it in workstreams/_candidates.md and ask for confirmation. |
| Time-bounded follow-up | Track it in next-actions/; use ongoing/ only when no next-meeting date is known. |
| Status review or handoff | Read workspace-summary.md, the portfolio, active actions, and open questions before reporting. |
Keep source records authoritative: summaries and workstreams link to meeting notes and inbox entries instead of duplicating them.
Track Work
- A workstream README owns its current status, owner, actions, and timeline; the portfolio is its index. Create a workstream only after its name and scope are confirmed.
- When a customer follow-up continues an existing question or delivered answer, keep the original exchange visible and append the follow-up chronologically in the owning deliverable or thread. If later evidence resolves an existing
要確認, append the resolution and evidence to that owning entry instead of creating a parallel source record. For month-bound_questionsledgers, append to the current month and reference the original entry instead of rewriting the prior month. - Put customer-agreed homework, self-initiated proposals, and supplementary research in separate
next-actions/folders. Task headers link back to their meeting source, while customer-shareable meeting tables exclude local task paths. - Use only
candidate,not-started,in-progress,blocked,done, ordropped. Before assigningnot-startedordone, inspect task-linked external work locations; workspace absence does not prove a deliverable is missing. When external progress and artifacts confirm completion, update the canonical action record and indexes in the same operation. Do not copy external artifacts unless the material lifecycle requires it. Ablockedrecord names the current owner and its transition condition. - Create
pj_{topic}/only when the topic spans multiple meetings and meets the detailed project-thread split conditions.
Meeting Notes Quality Gate
Before calling meeting notes done:
- Mark uncertain names, times, product names, prices, and support boundaries as
要確認; record the available transcript range and keep uncorroborated AI follow-ups provisional rather than assigning confirmed owners, deadlines, or deliverables. - Omit credentials and token-bearing image URLs from saved source text. After saving, ask for a Teams meeting/recap/recording URL if absent; record supplied links without asking again, and distinguish link capture from full-content verification.
- Extract open questions and work needing follow-up into
_questions/{YYYY-MM}.mdandnext-actions/in the same operation. - Keep local paths, internal links, internal speculation, and commercial terms out of customer-shareable content.
- Default to one working meeting note. Do not create
*_internal.mdsolely for unverified technical details; separate only on explicit request or when sensitive material cannot remain in a clearly marked internal section. - Customer highlights contain only confirmed items and split cross-side coordination into one
双方 (両者で調整)bucket. Apply Meeting Minutes Rules for details.
Done Criteria
- Setup: the managed scaffold exists and routing facts are captured.
- Operate: the canonical record, related question/action ledger, and relevant index reflect the input without duplicate source content.
- Review / Handoff: the report identifies authoritative sources, current owner, next action, and unresolved questions.
Key References
- Inbox Rules
- Meeting Minutes Rules
- Workspace Summary Rules
- Workstream Portfolio Rules
- Knowledge Ledger Rules
- Customer Material Lifecycle
- Project Thread Rules
Assets
Note:
assets/配下の prompt / instruction / template は、新しい顧客ワークスペースを初期化するときの コピー元 として使う scaffolding 用ファイル。ホスト workspace の.github/prompts/や.github/copilot-instructions.mdとは独立に進化させてよい(同期は必須ではない)。ホスト側で機能追加した場合に scaffolding にも反映したいときは、明示的にこのフォルダへ back-port する。
assets/_templates/: next-actions, knowledge ledger, attachments, and workspace templatesassets/*.prompt.md: inbox, meeting-note conversion, and question extraction prompts. Prompt files load only in the VS Code Local agent (not Agent Host sessions), so the routing rules incopilot-instructions.mdmust work without them.assets/copilot-instructions.md: generated workspace auto-routing rulesscripts/Test-ReceivedMaterialPlacement.ps1- Read-only root audit for unclassified received-material candidates
FAQ
What does the customer-workspace skill do?
Set up and operate customer workspaces: route customer updates, record meetings, track questions and actions, manage workstreams, and prepare handoffs. Use when creating or maintaining a customer workspace, processing meeting notes or updates, reviewing next actions, or preparing a handoff. Triggers: customer workspace, 顧客ワークスペース, inbox 追加, 議事録管理, 案件更新, 宿題管理, 引き継ぎ.
How do I install customer-workspace?
Run npx skills add https://github.com/aktsmm/agent-skills --skill customer-workspace in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/customer-workspace/ (personal) or .claude/skills/customer-workspace/ (project).
Which AI agents can use customer-workspace?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is customer-workspace safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains customer-workspace?
It is published by aktsmm in aktsmm/agent-skills. It has 57 recorded installs.
More from aktsmm/agent-skills
powerpoint-automation
aktsmm/agent-skills
Convert web articles or blog posts into slides through a content.json pipeline, translate or rebuild existing PPTX files onto a template, and edit an already-open PowerPoint deck with COM Automation (RefURL, overflow review). Use for article-to-slides conversion, deck translation, template-based rebuilds, or live COM editing of an open deck. Not for simply reading, inspecting, or lightly editing a .pptx file. Triggers on 記事をスライド化, スライド翻訳, 開いているパワポを編集, COM自動化, RefURL, content.json.
skill-finder
aktsmm/agent-skills
Search, install, and manage Agent Skills locally and from GitHub, then help decide whether the task really needs a skill or another customization primitive. Use when looking for skills, installing skills, managing a skill collection, or choosing between a skill, prompt, instruction, or agent.
ocr-super-surya
aktsmm/agent-skills
GPU-optimized OCR using Surya. Use when extracting text from images/screenshots, recovering image-only content in PDFs or slides, or inspecting multilingual document layouts. Preserve source locations and report recognition gaps separately from verified text.
code-simplifier
aktsmm/agent-skills
Open the skill to read its SKILL.md instructions.
book-writing-workspace
aktsmm/agent-skills
Operate a reusable technical book manuscript workspace with writing structure, reader persona SSOT, review rules, and optional Markdown to Re:VIEW/PDF support. Use when organizing a book manuscript repo, defining or reviewing target readers, standardizing chapter/section files, setting writing/review agents, handling publisher proof comments, or assessing an existing writing workspace. Triggers on book writing workspace, reader persona, reader audience, 読者ペルソナ, persona SSOT, technical book project, 執筆ワークスペース, manuscript workflow, Re:VIEW workspace, proof PDF, annotated proof, publisher proof comments, 校正, 校正注釈, and 初校.
vscode-extension-guide
aktsmm/agent-skills
Open the skill to read its SKILL.md instructions.
Popular ai & agents skills
google-agents-cli-publish
google/agents-cli
This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).
google-agents-cli-eval
google/agents-cli
This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).
skill-creator
anthropics/skills
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
codex-pet
prime-skills/runcomfy-agent-skills
Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.
full-output-enforcement
leonxlnx/taste-skill
Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.
executing-plans
obra/superpowers
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available