Vibecode Shade

You are building a lean indie version of Shade.

Create the following project files first, then implement the application by following them. Keep the files updated as decisions change. Do not collapse this into a single README or prompt.

===== README.md =====
# Shade indie build

## Goal

Build the smallest trustworthy replacement for the core Shade workflow for one developer or a tiny team.

## Scope

Walk my drives, pull keyframes and transcripts, embed both with CLIP and whisper into a local vector index, then search the whole library in plain English.

## Quick start

1. Install the documented dependencies.
2. Copy `.env.example` to `.env`.
3. Run the development command chosen during implementation.
4. Complete the acceptance checks in `BUILD_PLAN.md`.

## Honest limits

This build deliberately does not replace:
- cloud streaming of full-res files without downloading them first
- guest links with per-link permissions and roles
- built-in review, approval, and commenting
- face recognition and shot-type tagging that improves without your involvement
- team sync, so everyone searches the same index
- the NLE plugins and Slack integration

If those capabilities are essential, use Immich instead of pretending the gap is solved.

===== AGENTS.md =====
# Agent instructions

- Optimize for a working, understandable weekend build.
- Prefer the fewest moving parts that satisfy the brief.
- Do not invent cryptography, security guarantees, APIs, or compliance claims.
- Keep secrets out of source control and logs.
- Add focused tests for destructive, security-sensitive, and data-loss paths.
- Run the project checks before declaring the build complete.
- Record any deliberate shortcut in the README under "Tradeoffs".

===== BUILD_PLAN.md =====
# Build plan

## Original build brief

Build me a semantic search engine for my own footage to replace Shade. Requirements:

- A Python CLI plus a small FastAPI web UI on localhost. SQLite with sqlite-vec for
  the vectors and the metadata, one file at ~/FootageIndex/index.db.
- `index <folder>` walks the tree, and for every video ffmpeg pulls a keyframe every
  5 seconds plus one at each scene cut detected by the ffmpeg scene filter.
- Each keyframe is embedded with open_clip (ViT-B/32) and stored with its timestamp.
  Stills and photos get the same treatment as a single frame.
- Audio goes through whisper.cpp for a transcript with word timestamps, chunked into
  30-second windows and embedded with sentence-transformers for text search.
- The search box takes a plain sentence and searches image and transcript vectors
  together, returning ranked results as thumbnail, filename, and timecode. Clicking
  one opens the clip at that exact frame in a player.
- Indexing is incremental and resumable, keyed on file path plus mtime plus size, and
  prints a running count so an overnight run is checkable in the morning.
- Everything runs on my machine, models included · no accounts, no cloud, no
  telemetry, no API keys. Files are read only, never moved or renamed.
- Out of scope: face recognition, sharing links, review and comments, and team sync.
  Do not build auth or a server deployment, this is a single-user local tool.
- README: installing ffmpeg and whisper.cpp, first-run model downloads, an honest
  estimate of indexing hours per TB on CPU versus GPU, and how to reset the index.

## Required capabilities

- ffmpeg
- a CLIP model via transformers.js or Python
- whisper.cpp
- sqlite-vec or another local vector store
- a GPU, or patience measured in nights

## Delivery order

1. Scaffold the smallest runnable application and document its commands.
2. Implement the primary data model and core workflow.
3. Add validation, safe failure states, and persistence.
4. Cover the critical path with automated tests.
5. Exercise a clean install from the README and fix every missing step.

## Done when

- A new user can go from clone to first successful workflow using only the README.
- The core workflow works without paid infrastructure unless the brief requires it.
- Tests cover the highest-risk behavior.
- Known limitations are explicit rather than hidden.

===== .env.example =====
# Copy to .env and document every variable when it is introduced.
# Never put real credentials in this file.

APP_ENV=development
# Add only values required by the selected implementation.

You are building a lean indie version of Shade.

Create the following project files first, then implement the application by following them. Keep the files updated as decisions change. Do not collapse this into a single README or prompt.

===== README.md =====
# Shade indie build

## Goal

Build the smallest trustworthy replacement for the core Shade workflow for one developer or a tiny team.

## Scope

Walk my drives, pull keyframes and transcripts, embed both with CLIP and whisper into a local vector index, then search the whole library in plain English.

## Quick start

1. Install the documented dependencies.
2. Copy `.env.example` to `.env`.
3. Run the development command chosen during implementation.
4. Complete the acceptance checks in `BUILD_PLAN.md`.

## Honest limits

This build deliberately does not replace:
- cloud streaming of full-res files without downloading them first
- guest links with per-link permissions and roles
- built-in review, approval, and commenting
- face recognition and shot-type tagging that improves without your involvement
- team sync, so everyone searches the same index
- the NLE plugins and Slack integration

If those capabilities are essential, use Immich instead of pretending the gap is solved.

===== AGENTS.md =====
# Agent instructions

- Optimize for a working, understandable weekend build.
- Prefer the fewest moving parts that satisfy the brief.
- Do not invent cryptography, security guarantees, APIs, or compliance claims.
- Keep secrets out of source control and logs.
- Add focused tests for destructive, security-sensitive, and data-loss paths.
- Run the project checks before declaring the build complete.
- Record any deliberate shortcut in the README under "Tradeoffs".

===== BUILD_PLAN.md =====
# Build plan

## Original build brief

Build me a semantic search engine for my own footage to replace Shade. Requirements:

- A Python CLI plus a small FastAPI web UI on localhost. SQLite with sqlite-vec for
  the vectors and the metadata, one file at ~/FootageIndex/index.db.
- `index <folder>` walks the tree, and for every video ffmpeg pulls a keyframe every
  5 seconds plus one at each scene cut detected by the ffmpeg scene filter.
- Each keyframe is embedded with open_clip (ViT-B/32) and stored with its timestamp.
  Stills and photos get the same treatment as a single frame.
- Audio goes through whisper.cpp for a transcript with word timestamps, chunked into
  30-second windows and embedded with sentence-transformers for text search.
- The search box takes a plain sentence and searches image and transcript vectors
  together, returning ranked results as thumbnail, filename, and timecode. Clicking
  one opens the clip at that exact frame in a player.
- Indexing is incremental and resumable, keyed on file path plus mtime plus size, and
  prints a running count so an overnight run is checkable in the morning.
- Everything runs on my machine, models included · no accounts, no cloud, no
  telemetry, no API keys. Files are read only, never moved or renamed.
- Out of scope: face recognition, sharing links, review and comments, and team sync.
  Do not build auth or a server deployment, this is a single-user local tool.
- README: installing ffmpeg and whisper.cpp, first-run model downloads, an honest
  estimate of indexing hours per TB on CPU versus GPU, and how to reset the index.

## Required capabilities

- ffmpeg
- a CLIP model via transformers.js or Python
- whisper.cpp
- sqlite-vec or another local vector store
- a GPU, or patience measured in nights

## Delivery order

1. Scaffold the smallest runnable application and document its commands.
2. Implement the primary data model and core workflow.
3. Add validation, safe failure states, and persistence.
4. Cover the critical path with automated tests.
5. Exercise a clean install from the README and fix every missing step.

## Done when

- A new user can go from clone to first successful workflow using only the README.
- The core workflow works without paid infrastructure unless the brief requires it.
- Tests cover the highest-risk behavior.
- Known limitations are explicit rather than hidden.

===== .env.example =====
# Copy to .env and document every variable when it is introduced.
# Never put real credentials in this file.

APP_ENV=development
# Add only values required by the selected implementation.

You are building a production product version of Shade.

Create the following project files first, then implement the application by following them. Keep the files updated as decisions change. Do not collapse this into a single README or prompt.

===== PRODUCT.md =====
# Shade product brief

## Problem

The search half is real and rebuildable. Extract keyframes with ffmpeg, embed them with CLIP, transcribe the audio with whisper, put the vectors in SQLite, and 'the drone shot over the bridge at golden hour' finds the clip on your own drives. The open-source stack for that is mature and the result is genuinely good. What does not survive the port is what Shade has grown into: cloud streaming so an editor opens full-res without waiting on a download, per-link permissions and guest access, review and approval, and a model pipeline that keeps improving without you retraining anything. Solo, on local storage, the DIY version wins outright. On a team, you are rebuilding a platform and calling it a script.

## Product outcome

Walk my drives, pull keyframes and transcripts, embed both with CLIP and whisper into a local vector index, then search the whole library in plain English.

## Target user

A serious builder who needs a maintainable product foundation rather than a one-off demo.

## Required capabilities

- ffmpeg
- a CLIP model via transformers.js or Python
- whisper.cpp
- sqlite-vec or another local vector store
- a GPU, or patience measured in nights

## Explicit non-goals for v1

- cloud streaming of full-res files without downloading them first
- guest links with per-link permissions and roles
- built-in review, approval, and commenting
- face recognition and shot-type tagging that improves without your involvement
- team sync, so everyone searches the same index
- the NLE plugins and Slack integration

## Success criteria

- The primary workflow is measurable end to end.
- Setup is reproducible in a clean environment.
- Failure, recovery, and support paths are documented.
- Product claims match what the implementation actually guarantees.

===== ARCHITECTURE.md =====
# Architecture

## Starting brief

Build me a semantic search engine for my own footage to replace Shade. Requirements:

- A Python CLI plus a small FastAPI web UI on localhost. SQLite with sqlite-vec for
  the vectors and the metadata, one file at ~/FootageIndex/index.db.
- `index <folder>` walks the tree, and for every video ffmpeg pulls a keyframe every
  5 seconds plus one at each scene cut detected by the ffmpeg scene filter.
- Each keyframe is embedded with open_clip (ViT-B/32) and stored with its timestamp.
  Stills and photos get the same treatment as a single frame.
- Audio goes through whisper.cpp for a transcript with word timestamps, chunked into
  30-second windows and embedded with sentence-transformers for text search.
- The search box takes a plain sentence and searches image and transcript vectors
  together, returning ranked results as thumbnail, filename, and timecode. Clicking
  one opens the clip at that exact frame in a player.
- Indexing is incremental and resumable, keyed on file path plus mtime plus size, and
  prints a running count so an overnight run is checkable in the morning.
- Everything runs on my machine, models included · no accounts, no cloud, no
  telemetry, no API keys. Files are read only, never moved or renamed.
- Out of scope: face recognition, sharing links, review and comments, and team sync.
  Do not build auth or a server deployment, this is a single-user local tool.
- README: installing ffmpeg and whisper.cpp, first-run model downloads, an honest
  estimate of indexing hours per TB on CPU versus GPU, and how to reset the index.

## Boundaries

Separate the product into replaceable modules for interface, application logic, persistence, external integrations, and operational concerns. Keep domain logic independent from delivery frameworks and vendors.

## Production baseline

- Configuration: validated at startup with safe local defaults where possible.
- Security: least privilege, input validation, secret redaction, rate limits on abuse-prone paths, and no invented security primitives.
- Data: explicit schema and migrations, transactional writes where integrity matters, backup and restore instructions.
- Integrations: adapters around third-party providers, idempotent webhook or job processing, bounded retries, and timeouts.
- Observability: structured logs with request or operation IDs, an error-tracking hook, and health/readiness checks where a server exists.
- Quality: unit tests for domain rules, integration tests at module boundaries, and one end-to-end critical-path test.

## Decision records

For each major dependency, document why it was chosen, its failure mode, and how it can be replaced. Do not introduce infrastructure until a requirement justifies it.

===== AGENTS.md =====
# Agent instructions

- Read `PRODUCT.md` and `ARCHITECTURE.md` before changing code.
- Implement milestone by milestone; keep each change reviewable and leave the application runnable.
- Treat authentication, payments, encryption, imports, webhooks, and destructive actions as high-risk boundaries when present.
- Never invent cryptography or silently weaken a requirement to make a test pass.
- Use provider interfaces for external services and deterministic fakes in tests.
- Add migrations and rollback or recovery notes for persistent data changes.
- Log useful operational context without credentials, tokens, passwords, or personal data.
- Update documentation and run all checks before completing a milestone.

===== MILESTONES.md =====
# Delivery milestones

## M0 — Decisions and scaffold

- Confirm the runtime, persistence model, threat boundaries, and deployment target.
- Create a reproducible local environment and continuous checks.

## M1 — Core workflow

- Implement the smallest end-to-end product path with validation and tests.
- Keep integrations behind interfaces.

## M2 — Trust layer

- Add secure failure behavior, recovery paths, audit-relevant events, and data safeguards.
- Test abuse cases and destructive operations.

## M3 — Operability

- Add structured logs, error reporting hooks, health signals, backup/restore documentation, and deployment configuration.

## M4 — Release gate

- Run a clean-install test, critical-path end-to-end test, dependency review, and documented rollback exercise.
- Compare the shipped behavior with `PRODUCT.md` and publish remaining limitations.

===== OPERATIONS.md =====
# Operations

## Before release

- Validate configuration and secrets at startup.
- Define backup, restore, and rollback procedures and test them.
- Document logs, error tracking, health signals, and alert ownership.
- Set dependency update and vulnerability review expectations.

## Incident checklist

1. Contain the issue without destroying evidence or user data.
2. Record the timeline and affected scope.
3. Rotate exposed secrets and revoke compromised sessions or credentials.
4. Restore from a verified source when needed.
5. Document the root cause, remediation, and regression test.

## Launch constraint

Do not market omitted Shade capabilities as implemented. The v1 non-goals in `PRODUCT.md` remain user-visible limitations until they are deliberately delivered.

# Shade indie build

## Goal

Build the smallest trustworthy replacement for the core Shade workflow for one developer or a tiny team.

## Scope

Walk my drives, pull keyframes and transcripts, embed both with CLIP and whisper into a local vector index, then search the whole library in plain English.

## Quick start

1. Install the documented dependencies.
2. Copy `.env.example` to `.env`.
3. Run the development command chosen during implementation.
4. Complete the acceptance checks in `BUILD_PLAN.md`.

## Honest limits

This build deliberately does not replace:
- cloud streaming of full-res files without downloading them first
- guest links with per-link permissions and roles
- built-in review, approval, and commenting
- face recognition and shot-type tagging that improves without your involvement
- team sync, so everyone searches the same index
- the NLE plugins and Slack integration

If those capabilities are essential, use Immich instead of pretending the gap is solved.

# Agent instructions

- Optimize for a working, understandable weekend build.
- Prefer the fewest moving parts that satisfy the brief.
- Do not invent cryptography, security guarantees, APIs, or compliance claims.
- Keep secrets out of source control and logs.
- Add focused tests for destructive, security-sensitive, and data-loss paths.
- Run the project checks before declaring the build complete.
- Record any deliberate shortcut in the README under "Tradeoffs".

# Build plan

## Original build brief

Build me a semantic search engine for my own footage to replace Shade. Requirements:

- A Python CLI plus a small FastAPI web UI on localhost. SQLite with sqlite-vec for
  the vectors and the metadata, one file at ~/FootageIndex/index.db.
- `index <folder>` walks the tree, and for every video ffmpeg pulls a keyframe every
  5 seconds plus one at each scene cut detected by the ffmpeg scene filter.
- Each keyframe is embedded with open_clip (ViT-B/32) and stored with its timestamp.
  Stills and photos get the same treatment as a single frame.
- Audio goes through whisper.cpp for a transcript with word timestamps, chunked into
  30-second windows and embedded with sentence-transformers for text search.
- The search box takes a plain sentence and searches image and transcript vectors
  together, returning ranked results as thumbnail, filename, and timecode. Clicking
  one opens the clip at that exact frame in a player.
- Indexing is incremental and resumable, keyed on file path plus mtime plus size, and
  prints a running count so an overnight run is checkable in the morning.
- Everything runs on my machine, models included · no accounts, no cloud, no
  telemetry, no API keys. Files are read only, never moved or renamed.
- Out of scope: face recognition, sharing links, review and comments, and team sync.
  Do not build auth or a server deployment, this is a single-user local tool.
- README: installing ffmpeg and whisper.cpp, first-run model downloads, an honest
  estimate of indexing hours per TB on CPU versus GPU, and how to reset the index.

## Required capabilities

- ffmpeg
- a CLIP model via transformers.js or Python
- whisper.cpp
- sqlite-vec or another local vector store
- a GPU, or patience measured in nights

## Delivery order

1. Scaffold the smallest runnable application and document its commands.
2. Implement the primary data model and core workflow.
3. Add validation, safe failure states, and persistence.
4. Cover the critical path with automated tests.
5. Exercise a clean install from the README and fix every missing step.

## Done when

- A new user can go from clone to first successful workflow using only the README.
- The core workflow works without paid infrastructure unless the brief requires it.
- Tests cover the highest-risk behavior.
- Known limitations are explicit rather than hidden.

# Copy to .env and document every variable when it is introduced.
# Never put real credentials in this file.

APP_ENV=development
# Add only values required by the selected implementation.

# Shade product brief

## Problem

The search half is real and rebuildable. Extract keyframes with ffmpeg, embed them with CLIP, transcribe the audio with whisper, put the vectors in SQLite, and 'the drone shot over the bridge at golden hour' finds the clip on your own drives. The open-source stack for that is mature and the result is genuinely good. What does not survive the port is what Shade has grown into: cloud streaming so an editor opens full-res without waiting on a download, per-link permissions and guest access, review and approval, and a model pipeline that keeps improving without you retraining anything. Solo, on local storage, the DIY version wins outright. On a team, you are rebuilding a platform and calling it a script.

## Product outcome

Walk my drives, pull keyframes and transcripts, embed both with CLIP and whisper into a local vector index, then search the whole library in plain English.

## Target user

A serious builder who needs a maintainable product foundation rather than a one-off demo.

## Required capabilities

- ffmpeg
- a CLIP model via transformers.js or Python
- whisper.cpp
- sqlite-vec or another local vector store
- a GPU, or patience measured in nights

## Explicit non-goals for v1

- cloud streaming of full-res files without downloading them first
- guest links with per-link permissions and roles
- built-in review, approval, and commenting
- face recognition and shot-type tagging that improves without your involvement
- team sync, so everyone searches the same index
- the NLE plugins and Slack integration

## Success criteria

- The primary workflow is measurable end to end.
- Setup is reproducible in a clean environment.
- Failure, recovery, and support paths are documented.
- Product claims match what the implementation actually guarantees.

# Architecture

## Starting brief

Build me a semantic search engine for my own footage to replace Shade. Requirements:

- A Python CLI plus a small FastAPI web UI on localhost. SQLite with sqlite-vec for
  the vectors and the metadata, one file at ~/FootageIndex/index.db.
- `index <folder>` walks the tree, and for every video ffmpeg pulls a keyframe every
  5 seconds plus one at each scene cut detected by the ffmpeg scene filter.
- Each keyframe is embedded with open_clip (ViT-B/32) and stored with its timestamp.
  Stills and photos get the same treatment as a single frame.
- Audio goes through whisper.cpp for a transcript with word timestamps, chunked into
  30-second windows and embedded with sentence-transformers for text search.
- The search box takes a plain sentence and searches image and transcript vectors
  together, returning ranked results as thumbnail, filename, and timecode. Clicking
  one opens the clip at that exact frame in a player.
- Indexing is incremental and resumable, keyed on file path plus mtime plus size, and
  prints a running count so an overnight run is checkable in the morning.
- Everything runs on my machine, models included · no accounts, no cloud, no
  telemetry, no API keys. Files are read only, never moved or renamed.
- Out of scope: face recognition, sharing links, review and comments, and team sync.
  Do not build auth or a server deployment, this is a single-user local tool.
- README: installing ffmpeg and whisper.cpp, first-run model downloads, an honest
  estimate of indexing hours per TB on CPU versus GPU, and how to reset the index.

## Boundaries

Separate the product into replaceable modules for interface, application logic, persistence, external integrations, and operational concerns. Keep domain logic independent from delivery frameworks and vendors.

## Production baseline

- Configuration: validated at startup with safe local defaults where possible.
- Security: least privilege, input validation, secret redaction, rate limits on abuse-prone paths, and no invented security primitives.
- Data: explicit schema and migrations, transactional writes where integrity matters, backup and restore instructions.
- Integrations: adapters around third-party providers, idempotent webhook or job processing, bounded retries, and timeouts.
- Observability: structured logs with request or operation IDs, an error-tracking hook, and health/readiness checks where a server exists.
- Quality: unit tests for domain rules, integration tests at module boundaries, and one end-to-end critical-path test.

## Decision records

For each major dependency, document why it was chosen, its failure mode, and how it can be replaced. Do not introduce infrastructure until a requirement justifies it.

# Agent instructions

- Read `PRODUCT.md` and `ARCHITECTURE.md` before changing code.
- Implement milestone by milestone; keep each change reviewable and leave the application runnable.
- Treat authentication, payments, encryption, imports, webhooks, and destructive actions as high-risk boundaries when present.
- Never invent cryptography or silently weaken a requirement to make a test pass.
- Use provider interfaces for external services and deterministic fakes in tests.
- Add migrations and rollback or recovery notes for persistent data changes.
- Log useful operational context without credentials, tokens, passwords, or personal data.
- Update documentation and run all checks before completing a milestone.

# Delivery milestones

## M0 — Decisions and scaffold

- Confirm the runtime, persistence model, threat boundaries, and deployment target.
- Create a reproducible local environment and continuous checks.

## M1 — Core workflow

- Implement the smallest end-to-end product path with validation and tests.
- Keep integrations behind interfaces.

## M2 — Trust layer

- Add secure failure behavior, recovery paths, audit-relevant events, and data safeguards.
- Test abuse cases and destructive operations.

## M3 — Operability

- Add structured logs, error reporting hooks, health signals, backup/restore documentation, and deployment configuration.

## M4 — Release gate

- Run a clean-install test, critical-path end-to-end test, dependency review, and documented rollback exercise.
- Compare the shipped behavior with `PRODUCT.md` and publish remaining limitations.

# Operations

## Before release

- Validate configuration and secrets at startup.
- Define backup, restore, and rollback procedures and test them.
- Document logs, error tracking, health signals, and alert ownership.
- Set dependency update and vulnerability review expectations.

## Incident checklist

1. Contain the issue without destroying evidence or user data.
2. Record the timeline and affected scope.
3. Rotate exposed secrets and revoke compromised sessions or credentials.
4. Restore from a verified source when needed.
5. Document the root cause, remediation, and regression test.

## Launch constraint

Do not market omitted Shade capabilities as implemented. The v1 non-goals in `PRODUCT.md` remain user-visible limitations until they are deliberately delivered.

$ choose a build depth, inspect the files, then open the complete pack in your agent