# Send ClawHub Skill Publisher 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": "lum-skill-publisher", "name": "ClawHub Skill Publisher", "source": "tencent", "type": "skill", "category": "开发工具", "sourceUrl": "https://clawhub.ai/ragesaq/lum-skill-publisher", "canonicalUrl": "https://clawhub.ai/ragesaq/lum-skill-publisher", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/lum-skill-publisher", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=lum-skill-publisher", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "README.md", "SKILL.md", "examples/gap-analysis.md" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "lum-skill-publisher", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-05-01T14:45:40.301Z", "expiresAt": "2026-05-08T14:45:40.301Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=lum-skill-publisher", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=lum-skill-publisher", "contentDisposition": "attachment; filename=\"lum-skill-publisher-1.0.1.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "lum-skill-publisher" }, "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/lum-skill-publisher" }, "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/lum-skill-publisher", "downloadUrl": "https://openagent3.xyz/downloads/lum-skill-publisher", "agentUrl": "https://openagent3.xyz/skills/lum-skill-publisher/agent", "manifestUrl": "https://openagent3.xyz/skills/lum-skill-publisher/agent.json", "briefUrl": "https://openagent3.xyz/skills/lum-skill-publisher/agent.md" } } ``` ## Documentation ### ClawHub Skill Publisher v1 Turn a rough skill idea into a polished, publish-ready ClawHub listing — informed by what's actually working in the marketplace. Use this skill when you want to: Publish a new skill to ClawHub Audit an existing skill draft against marketplace standards Research what top-performing skills look like before writing yours ### Step 1 — Research Top Listings Install the most relevant published skills in a temp directory and read their SKILL.md + README.md: mkdir -p /tmp/ch-research # Search for skills in your category clawhub search "your-category-keyword" # Install top 3-5 results for analysis clawhub install --dir /tmp/ch-research --force clawhub install --dir /tmp/ch-research --force # (rate limit: add 3s sleep between installs) What to capture per skill: Description line: length, tone, value-first or feature-first? First sentence of SKILL.md: does it state the use case immediately? Structure: does it use tables, code blocks, headers? Word count (target: 400–700 words for SKILL.md) Sections present: commands, when-to-use, safety, version history Trust signals: safety section, version history, explicit opt-outs ### Step 2 — Gap Analysis Compare your draft against findings. Score each dimension: DimensionBest PracticeYour DraftActionDescription line≤160 chars, value-first, no buzzwords?Patch or OK"When to use"Explicit trigger + do/don't?Patch or OKCommands/interfaceSlash commands or trigger phrases?Patch or OKWord count (SKILL.md)400–700 words?Trim or expandTables vs. proseTables preferred for comparisons?Patch or OKVersion historyPresent, at bottom?Add or OKSafety sectionExplicit "never does X" list?Add or OKExamplesConcrete ✅/❌ pairs?Add or OKAttributionLink back to openclaw.ai / clawhub.ai?Add or OK ### Step 3 — Patch the Draft Apply gap findings. Priority order: Description line (most visible — fix first) "When to use" section (drives installs) Trim word count if over 700 (cut prose, keep tables) Add missing sections (safety, version history) Convert prose comparisons to tables Add examples file if none exists ### Step 4 — Publish # Verify auth clawhub whoami # Publish (run from workspace root or skill parent dir) clawhub publish ./skills/ \\ --slug \\ --name "Your Skill Name" \\ --version 1.0.0 \\ --changelog "Initial release" Published URL: https://clawhub.ai/skills/ ### Description Field (≤160 chars) The most important text. Shows in search results and install prompts. Formula: [What it does] + [how] + [key outcome]. ✅ Good: "Reduce AI costs by batching related asks into fewer responses. ~30–50% fewer API calls, no quality loss." ❌ Bad: "ClawSaver — Combines Linked Asks into Well-structured Sets for Affordable, Verified, Efficient Responses" ### SKILL.md Structure (what the agent reads) --- name: skill-name version: X.Y.Z description: "Same as listing description" metadata: {"openclaw":{"emoji":"🔧"}} --- # Skill Name vX > One-line positioning statement. [One paragraph: what it does and why.] ## When to Use [Use / Do not use — explicit conditions] ## Core Behavior / Commands [Tables preferred. Trigger phrases, commands, decision rules.] ## Safety [What it never does. Explicit opt-outs.] ## Installation [clawhub install command] ## Version History [- X.Y.Z — what changed] ### README.md Structure (humans + listing body) # Skill Name > Tagline ## Why [Skill Name]? [Problem → solution in 2-3 sentences] ## What It Does [Numbered or bulleted feature list] ## [Key Decision Table or Usage Example] ## Safety Model ## Installation ## Version ### What top skills have in common Value-first description (outcome before feature list) "When to use" is explicit — most top skills have do/don't lists Tables over prose for anything comparative Safety section is a trust signal — include it even if short Version history at the bottom — shows maintenance Word count 400–700 for SKILL.md; README can be longer ### What separates good from great Great: examples file with concrete ✅/❌ pairs Great: trigger phrase detection (tells agent when to activate) Great: explicit opt-outs ("say X to disable") Good but not great: long prose descriptions, missing opt-outs Avoid: backronyms or clever names in the description line (save for README) ### Category density (as of Feb 2026) Cost/token tracking: saturated — need a differentiated angle Batch/workflow: sparse — opportunity Provider-specific tools: mixed — Kimi-heavy, OpenAI moderate Productivity/meta-skills: sparse — opportunity ### File Checklist Before Publishing SKILL.md — frontmatter has name, version, description SKILL.md — word count 400–700 SKILL.md — has "When to Use" section SKILL.md — has Safety section SKILL.md — has Version History README.md — value-first, ≤600 words README.md — installation command correct examples/ — at least one example file (optional but recommended) Description line — ≤160 chars, value-first clawhub whoami — auth confirmed before publish ### Skill Type: Behavior-Change vs. Active Tool Most ClawHub skills are behavior-change skills — they work by shaping agent judgment through instructions, not by running code or intercepting requests at the system level. This is the same mechanism as execution-loop-breaker, token-saver, and most top listings. When writing a behavior-change skill: Be explicit in the description that it works through agent behavior, not automated interception Use language like "trains your agent to..." or "gives your agent the judgment to..." — not "automatically detects" or "intercepts" Don't overstate automation. "Teaches your agent to consolidate related asks" is honest. "Automatically batches requests" implies system-level routing that the skill doesn't do. The benefit is still real — behavior change produces real cost and efficiency improvements When a skill needs to be an active tool: Requires pre-response hooks or middleware (OpenClaw doesn't currently expose these) Requires script files (analyzer.js, optimizer.js) that actually run Example: a real token optimizer that reads context size and trims it before sending Bottom line: Instruction-based skills are legitimate and valuable. Just be honest about the scope. Users trust skills that set accurate expectations. ### Version History 1.0.1 — Added "Skill Type: Behavior-Change vs. Active Tool" lesson from ClawSaver development 1.0.0 — Initial release. Research workflow, gap analysis framework, listing anatomy, marketplace patterns from Feb 2026 analysis. ## Trust - Source: tencent - Verification: Indexed source record - Publisher: ragesaq - Version: 1.0.1 ## 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: 2026-05-01T14:45:40.301Z - Expires at: 2026-05-08T14:45:40.301Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/lum-skill-publisher) - [Send to Agent page](https://openagent3.xyz/skills/lum-skill-publisher/agent) - [JSON manifest](https://openagent3.xyz/skills/lum-skill-publisher/agent.json) - [Markdown brief](https://openagent3.xyz/skills/lum-skill-publisher/agent.md) - [Download page](https://openagent3.xyz/downloads/lum-skill-publisher)