Imagen AI

Build me a focused photo organization, culling and editing workflow for the personal core of Imagen AI. Requirements:

- Use Tauri 2, React, TypeScript, Rust image libraries, SQLite, and local filesystem access. Index selected folders without moving or rewriting originals and store only catalog data in SQLite.
- Paid product context: Apply reusable local adjustment presets and keep every edit reversible. Build only this DIY scope: Catalog local photos, generate previews, apply reusable non-destructive adjustment presets, keep every edit reversible, and export selected originals or rendered copies.
- Generate thumbnails and previews, read EXIF, detect duplicates by hash, and monitor file changes.
- Store edits as reversible parameters for crop, rotate, exposure, contrast, white balance, and saturation.
- Render exports to a new folder with explicit color space, quality, size, and metadata choices. Add catalog backup, missing-file repair, integrity scan, and a clear original-safety guarantee.
- Use a desktop window with source import, a work list, and result export. Required input or access: desktop computer; local photo library.
- Recovery: Keep the original media file untouched when decoding or rendering fails; show the failing job and allow retry.
- Acceptance: with one labelled sample, show the input, saved intermediate state, and exported result; verify this invariant: Original file hashes must remain unchanged; an interrupted render must leave no final file that looks complete.
- Out of scope: proprietary photographer-trained models, cloud processing, profiles, and workflow integration; proprietary raw-processing quality. Keep this a personal, inspectable workflow.
- Include a README with setup, a sample input, required keys or permissions, data location, and the supported scope.

# AGENTS.md — Build guide for Imagen AI

## Project scope
Catalog local photos, generate previews, apply reusable non-destructive adjustment presets, keep every edit reversible, and export selected originals or rendered copies.

Catalogue verdict: no. A consolation build is possible, but the paid product's decisive value sits outside a solo rebuild. For Imagen AI, apply reusable local adjustment presets and keep every edit reversible. The hard boundary is proprietary photographer-trained models, cloud processing, profiles, and workflow integration, plus image pipeline quality, models, and workflow polish.
Use the implementation prompt below to define the deliverable. Complete each phase's acceptance checks before extending the scope.

## Working agreement
- Inspect the repository and its existing instructions before choosing paths, dependencies or commands. Keep one coherent stack and explain changes to the proposed architecture.
- Plan a vertical slice that accepts a real input and produces the useful output described below. Persist only the state the prompt calls for; respect memory-only and upstream-managed workflows. Use fixtures only when they are clearly labelled.
- After scaffolding, document the actual install, development, check and build commands in README and keep them synchronized with the package or project manifest. Do not report commands as successful unless they ran.
- Work in small steps. At handoff, list implemented flows, checks actually performed, remaining blockers, and any credentials or provider setup the owner must supply.
- Do not publish, spend money, contact customers, delete source data or run irreversible migrations without the project owner's authorization.

## Prerequisites
- desktop computer
- local photo library
- sufficient disk space
- optional GPU for AI features

## Stack and architecture
- Use Tauri 2, React, TypeScript, Rust image libraries, SQLite, and local filesystem access.
- Index selected folders without moving or rewriting originals and store only catalog data in SQLite.
- Interface for this photo organization, culling and editing workflow: a desktop window with source import, a work list, and result export.

## Security and data integrity
- Keep the original media file untouched when decoding or rendering fails; show the failing job and allow retry.
- Validate imported records and file paths; keep user data within the storage location named in the prompt.
- State the practical limit: proprietary photographer-trained models, cloud processing, profiles, and workflow integration.

## Agent implementation rules
- Project rule, data: Index selected folders without moving or rewriting originals and store only catalog data in SQLite.
- Project rule, behavior: Generate thumbnails and previews, read EXIF, detect duplicates by hash, and monitor file changes.
- Project rule, recovery: Keep the original media file untouched when decoding or rendering fails; show the failing job and allow retry.

## Optional agent skills and references
No installable skill is listed for this guide; use the project rules and the stack documentation.

Read the linked SKILL.md and its dependencies before adding a skill. Select only the skills matching this project's runtime and task; their documentation does not supply API access, credentials or approval to perform external actions. Pin the reviewed revision where the tool supports it. Follow the chosen agent's documented project-level installation mechanism.

## Distribution ideas
These are optional planning notes. Obtain the owner's approval before publishing or contacting anyone.
- Demonstrate this photo organization, culling and editing operation with sample or user-owned data: Generate thumbnails and previews, read EXIF, detect duplicates by hash, and monitor file changes.
- Show a concrete result or export: Render exports to a new folder with explicit color space, quality, size, and metadata choices.
- State the practical limit: proprietary photographer-trained models, cloud processing, profiles, and workflow integration.

## Engineering roadmap
1. Phase 1, architecture and data: Use Tauri 2, React, TypeScript, Rust image libraries, SQLite, and local filesystem access. Index selected folders without moving or rewriting originals and store only catalog data in SQLite.
2. Phase 2, implement: Generate thumbnails and previews, read EXIF, detect duplicates by hash, and monitor file changes.
3. Phase 3, implement: Store edits as reversible parameters for crop, rotate, exposure, contrast, white balance, and saturation.
4. Phase 4, review and output: Render exports to a new folder with explicit color space, quality, size, and metadata choices.
5. Phase 5, recovery and acceptance: Keep the original media file untouched when decoding or rendering fails; show the failing job and allow retry. Verify this invariant with a saved fixture: Original file hashes must remain unchanged; an interrupted render must leave no final file that looks complete. State the practical limit: proprietary photographer-trained models, cloud processing, profiles, and workflow integration.

## Paid-product capabilities outside this build
- proprietary photographer-trained models, cloud processing, profiles, and workflow integration
- proprietary raw-processing quality
- cloud sync and sharing
- large AI models
- camera and print ecosystem

## Implementation prompt
Build me a focused photo organization, culling and editing workflow for the personal core of Imagen AI. Requirements:

- Use Tauri 2, React, TypeScript, Rust image libraries, SQLite, and local filesystem access. Index selected folders without moving or rewriting originals and store only catalog data in SQLite.
- Paid product context: Apply reusable local adjustment presets and keep every edit reversible. Build only this DIY scope: Catalog local photos, generate previews, apply reusable non-destructive adjustment presets, keep every edit reversible, and export selected originals or rendered copies.
- Generate thumbnails and previews, read EXIF, detect duplicates by hash, and monitor file changes.
- Store edits as reversible parameters for crop, rotate, exposure, contrast, white balance, and saturation.
- Render exports to a new folder with explicit color space, quality, size, and metadata choices. Add catalog backup, missing-file repair, integrity scan, and a clear original-safety guarantee.
- Use a desktop window with source import, a work list, and result export. Required input or access: desktop computer; local photo library.
- Recovery: Keep the original media file untouched when decoding or rendering fails; show the failing job and allow retry.
- Acceptance: with one labelled sample, show the input, saved intermediate state, and exported result; verify this invariant: Original file hashes must remain unchanged; an interrupted render must leave no final file that looks complete.
- Out of scope: proprietary photographer-trained models, cloud processing, profiles, and workflow integration; proprietary raw-processing quality. Keep this a personal, inspectable workflow.
- Include a README with setup, a sample input, required keys or permissions, data location, and the supported scope.

## Completion evidence
Demonstrate the prompt's acceptance scenarios against the scoped workflow. Include setup from a clean checkout and failure recovery. Check persistence across restart and export/restore only for the state the prompt says to store; for memory-only tools, confirm that temporary content is discarded as specified. Record actual results and remaining limitations. A detailed plan alone does not establish a working replacement.