Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Lovable MCP tool usage — deploy projects, message agent, test with agent-browser. Minimize credits by delegating to local repo.
.claude/skills/x-cmd-lovable/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-07 | ✗→✓ | ▲ Improved | 66% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 91% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 97% | 0% |
| case-12 | ✗→✓ | ▲ Improved | -21% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -7% | 0% |
claude mcp add --transport http lovable "https://mcp.lovable.dev"Full MCP reference: lovable-mcp-server.md.
local edit → npm run dev (verify) → git push → mcp__lovable__deploy_project → agent-browser testlist_projects query=<name> to confirm the correct project_id before any action. NEVER reuse an ID from memory or previous context. After deploy, wait 30s then verify the live site before reporting success.mcp__lovable__list_workspaces
mcp__lovable__list_projects workspace_id=<id> query=<name>
mcp__lovable__get_project project_id=<id>
mcp__lovable__deploy_project project_id=<id>
mcp__lovable__send_message project_id=<id> message="<instruction>" wait=true
mcp__lovable__get_diff project_id=<id> message_id=<msg_id>Quick overview: list_projects publish_status=published returns visitor counts per project. Export as TSV/CSV with schema: id, project, 24h, 7d, 30d, URL.
For detailed analytics (pageviews, bounce rate, traffic sources, devices), see ANALYTICS.md.
mcp__lovable__get_project_knowledge project_id=<id>
mcp__lovable__set_project_knowledge project_id=<id> content="<markdown>"Always get before set — replaces entirely. Max 10K chars. Also: get/set_workspace_knowledge.
agent-browser --profile /tmp/ab-default open <url> --session <proj>-mobile --headed
agent-browser --session <proj>-mobile set device "iPhone 14"
agent-browser --session <proj>-mobile snapshot -iFull usage: see skill0 agent-browser.
deploy_projectsend_message (high credit cost)agent-browser set device "iPhone 14"For PWA issues, sync problems, and more, see TROUBLESHOOTING.md.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 10,512 | 4,540 | -57% | 1 | 1 | 0% | 1,526 | 1,494 | -2% | 0 | 0 | — |
case-02 | fail→fail | 6,944 | 7,425 | +7% | 1 | 1 | 0% | 978 | 1,344 | +37% | 0 | 0 | — |
case-03 | fail→fail | 4,210 | 8,707 | +107% | 1 | 1 | 0% | 504 | 1,435 | +185% | 0 | 0 | — |
case-04 | pass→pass | 8,079 | 5,779 | -28% | 1 | 1 | 0% | 1,311 | 1,872 | +43% | 0 | 0 | — |
case-05 | pass→pass | 20,962 | 11,243 | -46% | 1 | 1 | 0% | 1,785 | 2,693 | +51% | 0 | 0 | — |
case-06 | fail→fail | 18,808 | 5,310 | -72% | 1 | 1 | 0% | 1,527 | 1,840 | +20% | 0 | 0 | — |
case-07 | fail→pass | 7,502 | 5,301 | -29% | 1 | 1 | 0% | 982 | 1,635 | +66% | 0 | 0 | — |
case-08 | pass→pass | 8,128 | 5,134 | -37% | 1 | 1 | 0% | 1,148 | 1,643 | +43% | 0 | 0 | — |
case-09 | fail→pass | 5,312 | 4,290 | -19% | 1 | 1 | 0% | 838 | 1,598 | +91% | 0 | 0 | — |
case-10 | fail→pass | 6,459 | 7,457 | +15% | 1 | 1 | 0% | 947 | 1,870 | +97% | 0 | 0 | — |
case-11 | pass→pass | 17,867 | 2,518 | -86% | 1 | 1 | 0% | 1,203 | 1,193 | -1% | 0 | 0 | — |
case-12 | fail→pass | 12,732 | 3,857 | -70% | 1 | 1 | 0% | 1,877 | 1,480 | -21% | 0 | 0 | — |
case-13 | fail→pass | 10,785 | 4,150 | -62% | 1 | 1 | 0% | 1,788 | 1,670 | -7% | 0 | 0 | — |
case-14 | fail→pass | 13,680 | 5,699 | -58% | 1 | 1 | 0% | 2,198 | 1,781 | -19% | 0 | 0 | — |
case-15 | fail→fail | 11,138 | 8,261 | -26% | 1 | 1 | 0% | 1,816 | 1,596 | -12% | 0 | 0 | — |
case-16 | fail→fail | 12,330 | 4,230 | -66% | 1 | 1 | 0% | 1,875 | 1,412 | -25% | 0 | 0 | — |
case-17 | pass→pass | 11,551 | 6,425 | -44% | 1 | 1 | 0% | 1,598 | 1,962 | +23% | 0 | 0 | — |
case-18 | fail→pass | 11,560 | 2,688 | -77% | 1 | 1 | 0% | 948 | 1,281 | +35% | 0 | 0 | — |
case-19 | fail→pass | 19,981 | 10,547 | -47% | 1 | 1 | 0% | 2,904 | 1,600 | -45% | 0 | 0 | — |
case-20 | fail→pass | 14,024 | 3,150 | -78% | 1 | 1 | 0% | 1,835 | 1,276 | -30% | 0 | 0 | — |
case-21 | fail→fail | 15,065 | 7,661 | -49% | 1 | 1 | 0% | 2,347 | 2,149 | -8% | 0 | 0 | — |
case-22 | fail→pass | 12,276 | 3,084 | -75% | 1 | 1 | 0% | 1,539 | 1,332 | -13% | 0 | 0 | — |
case-23 | pass→pass | 13,824 | 3,890 | -72% | 1 | 1 | 0% | 2,155 | 1,367 | -37% | 0 | 0 | — |
case-24 | fail→pass | 11,324 | 2,196 | -81% | 1 | 1 | 0% | 1,825 | 1,195 | -35% | 0 | 0 | — |
case-25 | fail→pass | 11,956 | 7,737 | -35% | 1 | 1 | 0% | 1,707 | 2,185 | +28% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 25 cases were attempted, and 23 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +48 percentage points is the difference between those two pass rates over the 23 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.