docs(13-01): complete content deduplication plan

Tasks completed: 2/2
- Add content hashing to save-generation API
- Add duplicate detection response handling

SUMMARY: .planning/phases/13-fix-duplicate-generations/13-01-SUMMARY.md

.planning/ROADMAP.md

@@ -157,16 +157,16 @@ Plans:
 Plans:
 - [x] 12-01: Extended cache TTL, fixed isImageInput, fixed stale node data in execution
 
-#### Phase 13: Fix Duplicate Generations
+#### Phase 13: Fix Duplicate Generations ✅
 
 **Goal**: Fix image deduplication - generations folder has duplicate images due to hashing failure
 **Depends on**: Phase 12
 **Research**: Likely (investigate current hashing implementation)
 **Research topics**: Current save-generation hashing logic, why duplicates are being created
-**Plans**: TBD
+**Plans**: 1 plan
 
 Plans:
-- [ ] 13-01: TBD (run /gsd:plan-phase 13 to break down)
+- [x] 13-01: Add MD5 content hashing and deduplication to save-generation API
 
 #### Phase 14: Fix Drag-Connect Node Creation Bugs
 
@@ -198,5 +198,5 @@ Phases execute in numeric order: 1 → 2 → 3 → 4 → 5 → 6 → 7 → 8 →
 | 10. Node Autosizing | v1.1 | 1/1 | Complete | 2026-01-12 |
 | 11. UI Polish | v1.1 | 1/1 | Complete | 2026-01-12 |
 | 12. Model Improvements | v1.1 | 1/1 | Complete | 2026-01-12 |
-| 13. Fix Duplicate Generations | v1.1 | 0/? | Not started | - |
+| 13. Fix Duplicate Generations | v1.1 | 1/1 | Complete | 2026-01-12 |
 | 14. Fix Drag-Connect Node Creation Bugs | v1.1 | 0/? | Not started | - |

.planning/STATE.md

@@ -5,22 +5,22 @@
 See: .planning/PROJECT.md (updated 2026-01-09)
 
 **Core value:** Provider infrastructure that dynamically discovers models from external APIs, enabling users to access hundreds of image/video generation models without hardcoding schemas.
-**Current focus:** Phase 13 - Fix Duplicate Generations
+**Current focus:** Phase 14 - Fix Drag-Connect Node Creation Bugs
 
 ## Current Position
 
-Phase: 12 of 14 (Model Improvements)
+Phase: 13 of 14 (Fix Duplicate Generations)
 Plan: 1 of 1 complete
 Status: Phase complete
-Last activity: 2026-01-12 - Completed Phase 12 fixes
+Last activity: 2026-01-12 - Completed 13-01-PLAN.md
 
-Progress: ██████░░░░ 50%
+Progress: ███████░░░ 55%
 
 ## Performance Metrics
 
 **Velocity:**
-- Total plans completed: 20
+- Total plans completed: 21
-- Average duration: 7.0 min
+- Average duration: 6.8 min
 - Total execution time: 2.4 hours
 
 **By Phase:**
@@ -38,9 +38,11 @@ Progress: ██████░░░░ 50%
 | 9. Video History | 1/1 | 12 min | 12 min |
 | 10. Node Autosizing | 1/1 | 2 min | 2 min |
 | 11. UI Polish | 1/1 | 8 min | 8 min |
+| 12. Model Improvements | 1/1 | - | - |
+| 13. Fix Duplicate Generations | 1/1 | 1 min | 1 min |
 
 **Recent Trend:**
-- Last 5 plans: 4 min, 14 min, 12 min, 2 min, 8 min
+- Last 5 plans: 14 min, 12 min, 2 min, 8 min, 1 min
 - Trend: Fast execution with established patterns
 
 ## Accumulated Context
@@ -89,6 +91,8 @@ Recent decisions affecting current work:
 - isImageInput() uses word-boundary checks (not substring) to avoid matching num_images
 - Workflow execution gets fresh node data from store (not stale captured array)
 - regenerateNode includes parameters in request body
+- MD5 content hashing for generation deduplication (fast, collision resistance not critical)
+- Hash prefix in filename for O(1) duplicate lookup
 
 ### Deferred Issues
 
@@ -109,6 +113,6 @@ Recent decisions affecting current work:
 ## Session Continuity
 
 Last session: 2026-01-12
-Stopped at: Phase 12 Model Improvements completed
+Stopped at: Phase 13 Fix Duplicate Generations completed
 Resume file: None
-Next action: Plan Phase 13 (Fix Duplicate Generations)
+Next action: Plan Phase 14 (Fix Drag-Connect Node Creation Bugs)

.planning/phases/13-fix-duplicate-generations/13-01-SUMMARY.md

@@ -0,0 +1,96 @@
+---
+phase: 13-fix-duplicate-generations
+plan: 01
+subsystem: api
+tags: [deduplication, md5, file-storage]
+
+# Dependency graph
+requires:
+  - phase: 06-video-and-polish
+    provides: save-generation API for images and videos
+provides:
+  - Content-based deduplication for generations folder
+  - Hash-prefixed filenames for O(1) duplicate lookup
+  - isDuplicate response field for caller awareness
+affects: []
+
+# Tech tracking
+tech-stack:
+  added: []
+  patterns: [content-hash-deduplication, hash-prefixed-filenames]
+
+key-files:
+  created: []
+  modified: [src/app/api/save-generation/route.ts]
+
+key-decisions:
+  - "MD5 for content hashing - fast for binary data, collision resistance not critical for dedup"
+  - "Hash prefix in filename - enables O(1) duplicate check without reading file contents"
+
+patterns-established:
+  - "Content deduplication via hash prefix in filename"
+
+issues-created: []
+
+# Metrics
+duration: 1min
+completed: 2026-01-12
+---
+
+# Phase 13 Plan 01: Add Content Deduplication Summary
+
+**MD5 content hashing prevents duplicate files in generations folder with hash-prefixed filenames for O(1) lookup**
+
+## Performance
+
+- **Duration:** 1 min
+- **Started:** 2026-01-12T10:03:51Z
+- **Completed:** 2026-01-12T10:04:52Z
+- **Tasks:** 2
+- **Files modified:** 1
+
+## Accomplishments
+
+- Added MD5 content hashing to save-generation API
+- Implemented duplicate detection by hash prefix lookup
+- Changed filename format to `{hash}_{prompt}.{ext}` for fast dedup checks
+- Added `isDuplicate` boolean to API response for caller awareness
+- Added deduplication event logging for observability
+
+## Task Commits
+
+Each task was committed atomically:
+
+1. **Task 1 & 2: Content hashing and duplicate detection** - `0e496bc` (feat)
+
+**Plan metadata:** (this commit)
+
+_Note: Tasks 1 and 2 were completed together as the implementation naturally combined both concerns._
+
+## Files Created/Modified
+
+- `src/app/api/save-generation/route.ts` - Added crypto import, hash computation, duplicate check, hash-prefixed filenames, isDuplicate response field
+
+## Decisions Made
+
+- **MD5 over SHA256:** MD5 is faster for binary data and collision resistance is not critical for local file deduplication (worst case is a rare false duplicate, not a security issue)
+- **Hash in filename:** Enables O(1) duplicate lookup via `files.find(f => f.startsWith(hash))` instead of reading/hashing all existing files
+
+## Deviations from Plan
+
+None - plan executed exactly as written.
+
+## Issues Encountered
+
+None
+
+## Next Phase Readiness
+
+- Phase 13 complete (single plan phase)
+- Deduplication active for all new generations
+- Existing files in generations folder (without hash prefix) will not be deduplicated against, but new files will deduplicate against each other
+- Ready for Phase 14: Fix Drag-Connect Node Creation Bugs
+
+---
+*Phase: 13-fix-duplicate-generations*
+*Completed: 2026-01-12*