# Send Workspace Standard to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - Download the package from Yavira. - Extract it into a folder your agent can access. - Paste one of the prompts below and point your agent at the extracted folder. ## Suggested prompts ### New install ```text I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete. ``` ### Upgrade existing ```text I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "workspace-standard", "name": "Workspace Standard", "source": "tencent", "type": "skill", "category": "效率提升", "sourceUrl": "https://clawhub.ai/marcus-daemon/workspace-standard", "canonicalUrl": "https://clawhub.ai/marcus-daemon/workspace-standard", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/workspace-standard", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=workspace-standard", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "README.md", "SKILL.md", "references/maintenance-checklist.md", "references/roles-guide.md", "scripts/workspace-audit.sh", "scripts/workspace-init.sh" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "workspace-standard", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-05-01T16:06:53.627Z", "expiresAt": "2026-05-08T16:06:53.627Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=workspace-standard", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=workspace-standard", "contentDisposition": "attachment; filename=\"workspace-standard-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "workspace-standard" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/workspace-standard" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/workspace-standard", "downloadUrl": "https://openagent3.xyz/downloads/workspace-standard", "agentUrl": "https://openagent3.xyz/skills/workspace-standard/agent", "manifestUrl": "https://openagent3.xyz/skills/workspace-standard/agent.json", "briefUrl": "https://openagent3.xyz/skills/workspace-standard/agent.md" } } ``` ## Documentation ### Workspace Standard A structured, portable workspace layout for OpenClaw. Gives your agent clear rules for where to put things, how to describe files, and how to keep memory under control. ### Why This Skill? If the user asks "why would I use this?" or "what does this do for me?": Without it: Files pile up in docs/, MEMORY.md bloats past its budget, the agent writes the right info to the wrong place, nothing is self-describing, and after a few weeks you can't tell a current reference from a stale plan. With it: Every file has a role and a place. MEMORY.md stays under budget. The audit script catches staleness and missing structure. New projects are one command. The agent knows where to write things without being told. It doesn't: Delete files, require API keys, or lock you in. Remove the skill and your files are still plain markdown. When explaining the value, run scripts/workspace-audit.sh on their workspace and show them what it finds. Concrete evidence beats abstract promises. ### 1. Install clawhub install workspace-standard ### 2. Bootstrap your workspace New workspace — creates all directories, seeds entity files, and sets up the project registry: bash skills/workspace-standard/scripts/workspace-init.sh Existing workspace — the skill also works if you already have files. Skip the bootstrap and go straight to step 3 (migration) or step 4 (audit). ### 3. Add your first project bash skills/workspace-standard/scripts/workspace-init.sh --project my-project This creates the project directory with README.md (including front-matter), standard subdirectories (references/, plans/, research/, reports/), and registers it in projects/_index.md. ### 4. Audit your workspace bash skills/workspace-standard/scripts/workspace-audit.sh Checks root files, MEMORY.md budget, directory structure, project health, front-matter coverage, staleness, and daily logs. Exit code = number of issues (0 = clean). ### 5. Customise (optional) Create .workspace-standard.yml in your workspace root to change defaults. See Configuration below. ### How the agent uses this skill Once installed, the agent automatically reads this skill when it needs to: Decide where to write something (the "Where to Write" table) Add a new project Run workspace maintenance Understand what a file's role is You don't need to tell the agent to use it — it triggers on matching tasks. ### Scripts ScriptPurposescripts/workspace-init.shBootstrap workspace or add a projectscripts/workspace-audit.shAudit workspace health and compliance # Full bootstrap (new workspace) bash skills/workspace-standard/scripts/workspace-init.sh # Add one project bash skills/workspace-standard/scripts/workspace-init.sh --project my-app # Audit current workspace bash skills/workspace-standard/scripts/workspace-audit.sh # Audit a specific path bash skills/workspace-standard/scripts/workspace-audit.sh /path/to/workspace ### Migrating an existing workspace If you already have a docs/ directory with files: Run the audit to see current state: bash scripts/workspace-audit.sh Create the structure: mkdir -p projects//{references,plans,research,reports} runbooks/ Categorise each file by role (use the decision tree below) Move with git mv to preserve history Add front-matter to moved files Update path references in AGENTS.md, MEMORY.md, and skills Trim MEMORY.md to budget (≤100 lines) Commit atomically ### The Role Taxonomy Every file gets a role — its job title. The role determines where the file lives, how it ages, and how the audit treats it. Read references/roles-guide.md for the full explanation with examples and tests for each role. RoleWhat it meansWhere it livesreferenceFacts about how things are right nowprojects/*/references/planHow you intend to do somethingprojects/*/plans/researchWhat you investigatedprojects/*/research/reportWhat you assessed at a point in timeprojects/*/reports/runbookHow to do something (procedure)runbooks/logWhat happenedmemory/entityStructured facts about a thingmemory/entities/ Quick decision tree: Is it about how things are? → reference. How to change them? → plan. Comparing options? → research. A snapshot assessment? → report. A reusable procedure? → runbook. What happened today? → log. A specific person/server/decision? → entity. ### ROLE Front-Matter Every substantive markdown file gets a YAML header: --- role: reference project: my-project # Omit for cross-project files status: current # active | current | completed | stale | archived created: 2026-02-01 updated: 2026-02-19 summary: "One-line description" --- Status lifecycle: active (being worked on) → current (living document) → stale (needs review) → archived (kept for history) ### Directory Layout workspace/ ├── MEMORY.md # Current state (≤100 lines) │ ├── memory/ # Episodic memory │ ├── YYYY-MM-DD.md # Daily logs (role: log) │ └── entities/ # People, servers, decisions (role: entity) │ ├── projects/ # Project-scoped work │ ├── _index.md # Project registry │ └── / │ ├── README.md # Overview + current state │ ├── references/ # role: reference │ ├── plans/ # role: plan │ ├── research/ # role: research │ └── reports/ # role: report │ ├── runbooks/ # Cross-project (role: runbook) │ ├── policies.md │ ├── lessons-learned.md │ └── / │ └── skills/ # Procedural memory └── /SKILL.md ### Where to Write What you learnedRoleWrite toFact about a projectreferenceprojects//references/How to fix somethingrunbookrunbooks/lessons-learned.mdOperational procedurerunbookskills/ or runbooks/What happened todaylogmemory/YYYY-MM-DD.mdCurrent state changed—MEMORY.mdPerson/server/decisionentitymemory/entities/Future work planplanprojects//plans/Research findingsresearchprojects//research/Audit or reviewreportprojects//reports/ ### MEMORY.md Budget MEMORY.md is loaded every session. Every line costs tokens. Budget: ≤100 lines / ~3500 tokens Contains: People, infrastructure, project pointers, lookup table, urgent items Never contains: History, lessons, architecture detail, completed items Over budget? Move detail to project files or runbooks, keep pointers ### Configuration Create .workspace-standard.yml in the workspace root to customise. All values are optional — defaults apply when omitted or when the file doesn't exist. budget: memory_lines: 100 # Max lines for MEMORY.md (default: 100) maintenance: stale_days: 14 # Days before flagging stale (default: 14) projects: subdirs: # Per-project subdirectories (default below) - references - plans - research - reports entities: # Seed files in memory/entities/ (default below) - people - servers - decisions ### New Project ./scripts/workspace-init.sh --project my-app Or manually: mkdir -p projects//{references,plans,research,reports} Create README.md with front-matter Add to projects/_index.md ### Migration (flat docs/ → structured) Create projects// and runbooks/ directories Categorise each file by role (use the decision tree above) Move with git mv (preserves history) Add front-matter to moved files Update path references in AGENTS.md, MEMORY.md, skills Trim MEMORY.md to budget Commit atomically ### Maintenance Run scripts/workspace-audit.sh weekly. See references/maintenance-checklist.md for the full procedure. Consolidate daily logs → extract facts to references, lessons to runbooks Prune MEMORY.md — remove resolved items, completed decisions Check front-matter — stale active/current files → verify or update Verify skills catalogue matches actual skills directory Commit maintenance changes ## Trust - Source: tencent - Verification: Indexed source record - Publisher: marcus-daemon - Version: 1.0.0 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-05-01T16:06:53.627Z - Expires at: 2026-05-08T16:06:53.627Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/workspace-standard) - [Send to Agent page](https://openagent3.xyz/skills/workspace-standard/agent) - [JSON manifest](https://openagent3.xyz/skills/workspace-standard/agent.json) - [Markdown brief](https://openagent3.xyz/skills/workspace-standard/agent.md) - [Download page](https://openagent3.xyz/downloads/workspace-standard)