$npx -y skills add prismatic-io/prismatic-skills --skill migration-frameworkCore knowledge for migrating integrations from other platforms (Boomi, Cyclr) to Prismatic Code Native Integrations. Provides the standard integration schema, schema-to-spec mapping, confidence scoring, and migration-aware code generation patterns. Use when the user says "migrate
| 1 | # Migration Framework |
| 2 | |
| 3 | ## Standard Integration Schema |
| 4 | |
| 5 | The intermediate JSON format that all platform parsers produce. See `references/standard-integration-schema.md`. |
| 6 | |
| 7 | Key sections: |
| 8 | - **metadata** — Source platform, confidence score, migration notes |
| 9 | - **integration** — Name, description, systems summary |
| 10 | - **flows** — Trigger type, step sequence with operations |
| 11 | - **api_profiles** — Request/response field definitions with nesting structure |
| 12 | - **systems** — Source and destination with connection details |
| 13 | - **data_transformations** — Field mappings and transformation functions |
| 14 | - **scripts** — Groovy scripts with full source code for TypeScript translation |
| 15 | - **error_handling** — Strategy (retry, fail, ignore) |
| 16 | - **config_variables** — Overrideable configuration values |
| 17 | - **migration_notes** — Manual review items, unsupported features |
| 18 | |
| 19 | ## Schema-to-Spec Mapping |
| 20 | |
| 21 | The `schema-to-answers` script maps schema fields to the integration YAML spec items. |
| 22 | See `references/schema-to-requirements-mapping.md`. |
| 23 | |
| 24 | **Pre-populated** (from parsed export — confirmed by user via proposal): |
| 25 | - `systems`, `trigger_type`, `schedule_value`, `data_flow` |
| 26 | - `source_system`, `destination_system` |
| 27 | - `transformations`, `error_handler_type`, `error_retry_*` |
| 28 | - `additional_requirements` (includes Groovy scripts for translation) |
| 29 | |
| 30 | **Left for live discovery** (requires platform interaction): |
| 31 | - `source_component`, `destination_component` — live registry search |
| 32 | - `*_connection_type`, `*_connection` — live connection search + user decision |
| 33 | - `flow_count`, `flow_definitions` — confirmed by user |
| 34 | |
| 35 | ## Confidence Scoring |
| 36 | |
| 37 | Per-element confidence: |
| 38 | - **high** (0.8-1.0): Direct 1:1 mapping, deterministic |
| 39 | - **medium** (0.5-0.79): Requires interpretation or custom code |
| 40 | - **low** (0.0-0.49): No direct equivalent, manual design needed |
| 41 | |
| 42 | Overall = weighted average. Drives reviewer focus. |
| 43 | |
| 44 | ## Migration-Aware Code Generation |
| 45 | |
| 46 | When `migration-schema.json` exists in the session, `code-plan` emits `<migration-context>`: |
| 47 | - **api_profiles** — exact field names and nesting paths. Use these, don't invent field names. |
| 48 | - **script_translations** — full Groovy source with input/output contracts. Translate completely. |
| 49 | - **field_mappings** — source→destination field mappings from transformations. |
| 50 | - **endpoints** — known API paths with confidence scores. |
| 51 | |
| 52 | See `references/migration-code-gen-guide.md` for translation patterns. |
| 53 | |
| 54 | ## Key References |
| 55 | |
| 56 | - `references/standard-integration-schema.md` — Full schema definition with examples |
| 57 | - `references/schema-to-requirements-mapping.md` — Spec item ID mapping |
| 58 | - `references/migration-code-gen-guide.md` — API profiles, script translation, field mapping patterns |