Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Next.js development tooling via MCP. Inspect routes, components, build info, and debug Next.js apps. Use when working on Next.js applications, debugging routing, or inspecting app structure. NOT for general React or non-Next.js projects.
.claude/skills/aiskillstore-nextjs-devtools/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -34% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -33% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -44% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -54% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -37% | 0% |
Inspect and debug Next.js applications via MCP server.
bash# Start server (spawns on-demand) bash scripts/start-server.sh # Or use directly via mcp-client python3 scripts/mcp-client.py call \ -s "npx next-devtools-mcp@latest" \ -t list-routes
| Tool | Description | |------|-------------| | list-routes | Get all app routes | | get-route-info | Details for specific route | | list-components | React components in app | | get-build-info | Build configuration | | get-config | next.config.js settings |
bash# List all routes python3 scripts/mcp-client.py call \ -s "npx next-devtools-mcp@latest" \ -t list-routes # Get specific route details python3 scripts/mcp-client.py call \ -s "npx next-devtools-mcp@latest" \ -t get-route-info \ -p '{"route": "/api/auth"}'
bash# Check build config python3 scripts/mcp-client.py call \ -s "npx next-devtools-mcp@latest" \ -t get-build-info # Check next.config.js python3 scripts/mcp-client.py call \ -s "npx next-devtools-mcp@latest" \ -t get-config
bashpython3 scripts/mcp-client.py call \ -s "npx next-devtools-mcp@latest" \ -t list-components
For persistent server (multiple calls):
bashbash scripts/start-server.sh # Server runs on default port # Use mcp-client.py with -u flag instead of -s
For single calls, use -s flag which spawns server per-call:
bashpython3 scripts/mcp-client.py call \ -s "npx next-devtools-mcp@latest" \ -t <tool-name>
| Issue | Solution | |-------|----------| | Server not starting | Check npx next-devtools-mcp@latest works manually | | No routes found | Ensure running from Next.js project root | | Build info empty | Run next build first |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 11,779 | 2,792 | -76% | 1 | 1 | 0% | 1,784 | 1,178 | -34% | 0 | 0 | — |
case-02 | fail→pass | 11,607 | 3,967 | -66% | 1 | 1 | 0% | 1,900 | 1,274 | -33% | 0 | 0 | — |
case-03 | fail→pass | 11,266 | 2,784 | -75% | 1 | 1 | 0% | 1,971 | 1,098 | -44% | 0 | 0 | — |
case-04 | fail→pass | 14,018 | 2,806 | -80% | 1 | 1 | 0% | 2,448 | 1,135 | -54% | 0 | 0 | — |
case-05 | pass→pass | 5,584 | 2,830 | -49% | 1 | 1 | 0% | 962 | 1,079 | +12% | 0 | 0 | — |
case-06 | pass→pass | 10,542 | 2,541 | -76% | 1 | 1 | 0% | 1,639 | 1,022 | -38% | 0 | 0 | — |
case-07 | pass→pass | 8,404 | 3,402 | -60% | 1 | 1 | 0% | 1,377 | 1,251 | -9% | 0 | 0 | — |
case-08 | fail→pass | 11,369 | 3,719 | -67% | 1 | 1 | 0% | 2,022 | 1,266 | -37% | 0 | 0 | — |
case-09 | pass→pass | 12,794 | 2,241 | -82% | 1 | 1 | 0% | 2,379 | 1,047 | -56% | 0 | 0 | — |
case-10 | fail→pass | 6,573 | 1,788 | -73% | 1 | 1 | 0% | 1,150 | 954 | -17% | 0 | 0 | — |
case-11 | pass→pass | 6,852 | 2,084 | -70% | 1 | 1 | 0% | 1,266 | 1,016 | -20% | 0 | 0 | — |
case-12 | fail→pass | 11,784 | 1,322 | -89% | 1 | 1 | 0% | 1,926 | 849 | -56% | 0 | 0 | — |
case-13 | fail→pass | 12,334 | 1,788 | -86% | 1 | 1 | 0% | 2,294 | 957 | -58% | 0 | 0 | — |
case-14 | pass→pass | 7,002 | 3,069 | -56% | 1 | 1 | 0% | 1,264 | 1,090 | -14% | 0 | 0 | — |
case-15 | fail→pass | 10,431 | 4,902 | -53% | 1 | 1 | 0% | 1,668 | 1,334 | -20% | 0 | 0 | — |
case-16 | fail→pass | 7,939 | 2,802 | -65% | 1 | 1 | 0% | 1,244 | 1,091 | -12% | 0 | 0 | — |
case-17 | fail→pass | 6,908 | 1,906 | -72% | 1 | 1 | 0% | 1,094 | 960 | -12% | 0 | 0 | — |
case-18 | fail→pass | 7,218 | 3,122 | -57% | 1 | 1 | 0% | 1,106 | 1,166 | +5% | 0 | 0 | — |
case-19 | fail→pass | 6,807 | 1,991 | -71% | 1 | 1 | 0% | 1,094 | 957 | -13% | 0 | 0 | — |
case-20 | pass→pass | 25,631 | 5,451 | -79% | 1 | 1 | 0% | 2,289 | 1,376 | -40% | 0 | 0 | — |
case-21 | fail→pass | 7,744 | 6,958 | -10% | 1 | 1 | 0% | 1,275 | 1,957 | +53% | 0 | 0 | — |
case-22 | fail→pass | 7,783 | 4,440 | -43% | 1 | 1 | 0% | 1,335 | 1,347 | +1% | 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. 22 cases were attempted. The headline lift of +68 percentage points is the difference between those two pass rates over the 22 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.