Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Install, verify, troubleshoot, and operate the Adobe Premiere Pro MCP server. Use when a user wants an agent to set up Premiere MCP, connect Claude Code/Codex/Claude Desktop, control Premiere, import media, build sequences, edit timelines, apply effects, or diagnose bridge issues.
.claude/skills/hetpatel-11-premiere-pro-mcp/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 19% | 0% |
Use this skill when working with the Adobe Premiere Pro MCP server from hetpatel-11/Adobe_Premiere_Pro_MCP.
tools/list is a small always-on set. Call search_tools (BM25 query or regex pattern), then invoke_tool with the exact name. PREMIERE_MCP_TOOLSET=full lists every tool.get_project_info, list_sequences, list_project_items, get_active_sequence, and relevant resource reads.import_media, then place the imported project item IDs on a sequence./tmp/premiere-mcp-bridge unless the user explicitly configured another path.success: false, report the exact error and run diagnostics before retrying blindly.If the user asks you to install or set up the MCP:
bashgit clone https://github.com/hetpatel-11/Adobe_Premiere_Pro_MCP.git cd Adobe_Premiere_Pro_MCP
bashnpm run setup:mac
bashnpm install npm run build
textcommand: node /absolute/path/to/Adobe_Premiere_Pro_MCP/dist/index.js env: PREMIERE_TEMP_DIR=/tmp/premiere-mcp-bridge
For Codex, prefer:
bashcodex mcp add premiere_pro --env PREMIERE_TEMP_DIR=/tmp/premiere-mcp-bridge -- node /absolute/path/to/Adobe_Premiere_Pro_MCP/dist/index.js
After installing:
Window > Extensions > MCP Bridge (CEP).Temp Directory to /tmp/premiere-mcp-bridge.Save Configuration.Start Bridge.If Premiere is not running, you can install/build/register the MCP, but tell the user live tool verification requires Premiere and the CEP bridge panel.
Run local checks:
bashnpm run setup:doctor
If Premiere is running and the bridge is started, verify with safe read-only calls:
get_project_infolist_sequenceslist_project_itemsFor deeper validation in a disposable project, create a test sequence with create_sequence_from_clips or create_sequence plus a real .sqpreset path, then call list_sequences and confirm it exists. Do not call blank sequence creation without a preset because newer Premiere versions can open a native dialog that blocks CEP.
create_sequence_from_clips so Premiere derives settings without a native dialog. Use create_sequence only with a real .sqpreset; use duplicate_sequence with clearContents=true when an existing sequence defines the intended settings.assemble_product_spot or build_brand_spot_from_mogrt_and_assets when the user's request fits those workflows.apply_effect with Black & White rather than generic saturation-only changes.sequenceId when available.If commands time out or report bridge errors:
Window > Extensions > MCP Bridge (CEP) is open and bridge is started.bashnpm run setup:doctor
Run Diagnostics in the CEP panel./tmp/premiere-mcp-bridge/premiere-mcp-diagnostics-latest.json if it exists.Common fixes:
ENOENT on temp directory: create /tmp/premiere-mcp-bridge, save config again, restart bridge.list_sequences or the relevant list tool to confirm state before retrying.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-22 | pass→pass | 32,203 | 19,519 | -39% | 1 | 1 | 0% | 2,666 | 3,661 | +37% | 0 | 0 | — |
case-01 | fail→pass | 16,473 | 16,346 | -1% | 1 | 1 | 0% | 2,660 | 3,491 | +31% | 0 | 0 | — |
case-02 | fail→fail | 10,209 | 20,342 | +99% | 1 | 1 | 0% | 1,701 | 1,831 | +8% | 0 | 0 | — |
case-03 | fail→fail | 13,863 | 5,579 | -60% | 1 | 1 | 0% | 2,230 | 1,612 | -28% | 0 | 0 | — |
case-04 | pass→pass | 13,529 | 10,789 | -20% | 1 | 1 | 0% | 2,277 | 3,307 | +45% | 0 | 0 | — |
case-05 | fail→pass | 15,661 | 8,894 | -43% | 1 | 1 | 0% | 1,732 | 2,035 | +17% | 0 | 0 | — |
case-06 | fail→pass | 20,486 | 11,486 | -44% | 1 | 1 | 0% | 2,246 | 2,276 | +1% | 0 | 0 | — |
case-07 | fail→pass | 16,839 | 7,368 | -56% | 1 | 1 | 0% | 1,855 | 2,632 | +42% | 0 | 0 | — |
case-08 | fail→pass | 17,611 | 6,435 | -63% | 1 | 1 | 0% | 1,986 | 2,357 | +19% | 0 | 0 | — |
case-09 | fail→fail | 10,511 | 18,549 | +76% | 1 | 1 | 0% | 1,052 | 1,858 | +77% | 0 | 0 | — |
case-16 | fail→fail | 9,793 | 11,896 | +21% | 1 | 1 | 0% | 793 | 1,577 | +99% | 0 | 0 | — |
case-10 | fail→fail | 3,905 | 8,134 | +108% | 1 | 1 | 0% | 586 | 1,793 | +206% | 0 | 0 | — |
case-11 | fail→pass | 10,583 | 12,439 | +18% | 1 | 1 | 0% | 1,559 | 2,489 | +60% | 0 | 0 | — |
case-12 | fail→fail | 7,388 | 15,627 | +112% | 1 | 1 | 0% | 1,120 | 1,667 | +49% | 0 | 0 | — |
case-13 | fail→pass | 10,889 | 10,839 | -0% | 1 | 1 | 0% | 1,715 | 2,186 | +27% | 0 | 0 | — |
case-14 | fail→pass | 20,048 | 15,024 | -25% | 1 | 1 | 0% | 2,296 | 3,046 | +33% | 0 | 0 | — |
case-15 | fail→pass | 18,332 | 2,518 | -86% | 1 | 1 | 0% | 1,660 | 1,628 | -2% | 0 | 0 | — |
case-17 | fail→pass | 16,352 | 6,068 | -63% | 1 | 1 | 0% | 2,599 | 2,313 | -11% | 0 | 0 | — |
case-18 | fail→pass | 17,121 | 6,570 | -62% | 1 | 1 | 0% | 1,841 | 2,309 | +25% | 0 | 0 | — |
case-19 | pass→pass | 11,056 | 3,508 | -68% | 1 | 1 | 0% | 1,892 | 1,911 | +1% | 0 | 0 | — |
case-20 | fail→pass | 22,940 | 13,513 | -41% | 1 | 1 | 0% | 2,861 | 2,775 | -3% | 0 | 0 | — |
case-21 | pass→pass | 14,995 | 23,034 | +54% | 1 | 1 | 0% | 1,522 | 2,818 | +85% | 0 | 0 | — |
case-23 | pass→pass | 9,994 | 13,309 | +33% | 1 | 1 | 0% | 1,481 | 2,801 | +89% | 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. 23 cases were attempted, and 17 counted toward the lift figure. The other 6 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 +52 percentage points is the difference between those two pass rates over the 17 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/30/2026 | +52% |
Other measured skills in the registry, with their headline benchmark lift.