Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Postman concepts and MCP tool guidance. Loaded when working with Postman MCP tools to make better decisions about tool selection and workarounds.
.claude/skills/kunanonj-cursor-plugin-postman-postman-knowledge/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -10% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 58% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -22% | 0% |
Core Postman concepts and MCP tool guidance for making better decisions when working with Postman APIs.
| Goal | Approach | |------|----------| | Push code changes to Postman | Spec Hub + sync (createSpec -> syncCollectionWithSpec) | | Consume a Postman API in code | Read collection + codegen (getCollection -> generate client) | | Find an API | Search private network (searchPostmanElementsInPrivateNetwork) or user's workspace (getWorkspaces + getCollections) or public network (searchPostmanElementsInPublicNetwork) | | Test an API | Run collection (runCollection with environment) | | Fake an API for frontend | Mock server (createMock from collection with examples) | | Document an API | Analyze collection completeness, fill gaps, improve descriptions | | Secure an API | Audit spec and collection against OWASP API Top 10 |
getWorkspaces -- List all accessible workspacesgetCollections -- List collections in a workspacegetCollection -- Get full collection detail (always use full model)getCollectionFolder -- Get folder contentsgetCollectionRequest -- Get request details (method, URL, body, params)getCollectionResponse -- Get saved example responsesgetEnvironment / getEnvironments -- Get environment variablesgetSpecDefinition -- Get the full OpenAPI/AsyncAPI specgetAllSpecs -- List all specs in a workspacecreateCollection -- Create new collection (cannot nest folders, use decomposed approach)createCollectionFolder -- Create folder in collectioncreateCollectionRequest -- Create request in collection or foldercreateCollectionResponse -- Save example responsecreateEnvironment -- Create environment with variablescreateSpec -- Upload OpenAPI spec to Spec HubupdateSpecFile -- Update an existing specupdateCollectionRequest -- Modify request detailscreateMock -- Create mock server from collectionsyncCollectionWithSpec -- Update collection from spec (async, OpenAPI 3.0 only)syncSpecWithCollection -- Update spec from collection changesgenerateCollection -- Generate collection from spec (async)runCollection -- Execute collection testssearchPostmanElementsInPrivateNetwork -- Search private/org APIs (default)searchPostmanElementsInPublicNetwork -- Search public Postman networkgetTaggedEntities -- Find by tagFor detailed limitations and workarounds, see the companion file mcp-limitations.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 | 11,900 | 5,404 | -55% | 1 | 1 | 0% | 1,813 | 2,058 | +14% | 0 | 0 | — |
case-02 | fail→pass | 12,818 | 5,109 | -60% | 1 | 1 | 0% | 2,049 | 1,843 | -10% | 0 | 0 | — |
case-03 | fail→fail | 5,580 | 3,102 | -44% | 1 | 1 | 0% | 999 | 1,485 | +49% | 0 | 0 | — |
case-04 | fail→pass | 6,406 | 5,426 | -15% | 1 | 1 | 0% | 1,275 | 2,010 | +58% | 0 | 0 | — |
case-05 | fail→pass | 5,548 | 12,159 | +119% | 1 | 1 | 0% | 989 | 1,342 | +36% | 0 | 0 | — |
case-06 | pass→pass | 4,063 | 2,979 | -27% | 1 | 1 | 0% | 724 | 1,506 | +108% | 0 | 0 | — |
case-07 | fail→pass | 11,045 | 4,709 | -57% | 1 | 1 | 0% | 1,772 | 1,719 | -3% | 0 | 0 | — |
case-08 | fail→pass | 9,554 | 2,112 | -78% | 1 | 1 | 0% | 1,678 | 1,308 | -22% | 0 | 0 | — |
case-09 | pass→pass | 5,080 | 1,388 | -73% | 1 | 1 | 0% | 904 | 1,173 | +30% | 0 | 0 | — |
case-10 | fail→pass | 9,357 | 2,891 | -69% | 1 | 1 | 0% | 1,831 | 1,382 | -25% | 0 | 0 | — |
case-11 | fail→pass | 9,176 | 1,666 | -82% | 1 | 1 | 0% | 1,556 | 1,261 | -19% | 0 | 0 | — |
case-12 | fail→pass | 11,381 | 1,756 | -85% | 1 | 1 | 0% | 2,203 | 1,286 | -42% | 0 | 0 | — |
case-13 | fail→pass | 8,997 | 2,035 | -77% | 1 | 1 | 0% | 1,642 | 1,289 | -21% | 0 | 0 | — |
case-14 | fail→pass | 8,553 | 1,770 | -79% | 1 | 1 | 0% | 1,614 | 1,267 | -21% | 0 | 0 | — |
case-15 | fail→pass | 5,004 | 2,436 | -51% | 1 | 1 | 0% | 929 | 1,442 | +55% | 0 | 0 | — |
case-16 | fail→pass | 8,203 | 1,953 | -76% | 1 | 1 | 0% | 1,403 | 1,266 | -10% | 0 | 0 | — |
case-17 | fail→pass | 7,732 | 3,050 | -61% | 1 | 1 | 0% | 1,551 | 1,410 | -9% | 0 | 0 | — |
case-18 | fail→pass | 7,756 | 1,926 | -75% | 1 | 1 | 0% | 1,404 | 1,322 | -6% | 0 | 0 | — |
case-19 | fail→pass | 7,627 | 2,875 | -62% | 1 | 1 | 0% | 1,453 | 1,411 | -3% | 0 | 0 | — |
case-20 | fail→pass | 3,002 | 3,357 | +12% | 1 | 1 | 0% | 572 | 1,593 | +178% | 0 | 0 | — |
case-21 | pass→pass | 1,738 | 1,735 | -0% | 1 | 1 | 0% | 363 | 1,277 | +252% | 0 | 0 | — |
case-22 | pass→pass | 3,045 | 2,292 | -25% | 1 | 1 | 0% | 518 | 1,355 | +162% | 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 +73 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.