Elicit

Build me a focused AI research assistant workflow for the personal core of Elicit. Requirements:

- Use exactly this stack: Next.js 15 + TypeScript + SQLite + Playwright. Model source URLs, saved items, reading state, annotations, and exports; keep stable source IDs and timestamps.
- Paid product context: Finds, screens, and extracts evidence from research papers. Build only this DIY scope: Build a private AI research assistant workspace that imports user-supplied URLs or files, extracts metadata, supports notes, and searches the local corpus.
- For AI research assistant, accept a user-supplied URL or file and save its original URL, fetch time, and content hash.
- Extract readable content, deduplicate imports, and preserve notes or highlights by stable source position.
- Search the saved collection and export an item with its source metadata.
- Use a local web page with input, progress, review, and export views. Required input or access: browser automation for user-authorized imports; optional OpenAI API key stored in .env. Keep credentials in .env.
- Acceptance: with one labelled sample, show the input, saved intermediate state, and exported result; verify this invariant: An unreachable URL keeps its bookmark and error state; duplicate imports must not discard existing highlights.
- Out of scope: citation graph scale; team libraries and institutional access. 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 Elicit

## Project scope
Build a private AI research assistant workspace that imports user-supplied URLs or files, extracts metadata, supports notes, and searches the local corpus.

Catalogue verdict: no. Do not mistake the interface for the product. Elicit's durable value is data, import reliability, which a solo one-shot build cannot reproduce responsibly. The prompt therefore builds only the closest honest personal consolation tool.
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
- Node.js 22
- browser automation for user-authorized imports
- optional OpenAI API key stored in .env
- Explicit README warning that this is a consolation build, not a production replacement

## Stack and architecture
- Use exactly this stack: Next.js 15 + TypeScript + SQLite + Playwright.
- Model source URLs, saved items, reading state, annotations, and exports; keep stable source IDs and timestamps.
- Interface for this AI research assistant workflow: a local web page with input, progress, review, and export views.

## Security and data integrity
- An unreachable URL keeps its bookmark and error state; duplicate imports must not discard existing highlights.
- Keep API keys and OAuth secrets in .env; redact them from logs, generated files, and error responses.
- State the practical limit: citation graph scale.

## Agent implementation rules
- Project rule, data: Model source URLs, saved items, reading state, annotations, and exports; keep stable source IDs and timestamps.
- Project rule, behavior: For AI research assistant, accept a user-supplied URL or file and save its original URL, fetch time, and content hash.
- Project rule, recovery: An unreachable URL keeps its bookmark and error state; duplicate imports must not discard existing highlights.

## 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 AI research assistant operation with sample or user-owned data: For AI research assistant, accept a user-supplied URL or file and save its original URL, fetch time, and content hash.
- Show a concrete result or export: Search the saved collection and export an item with its source metadata.
- State the practical limit: citation graph scale.

## Engineering roadmap
1. Phase 1, architecture and data: Use exactly this stack: Next.js 15 + TypeScript + SQLite + Playwright. Model source URLs, saved items, reading state, annotations, and exports; keep stable source IDs and timestamps.
2. Phase 2, implement: For AI research assistant, accept a user-supplied URL or file and save its original URL, fetch time, and content hash.
3. Phase 3, implement: Extract readable content, deduplicate imports, and preserve notes or highlights by stable source position.
4. Phase 4, review and output: Search the saved collection and export an item with its source metadata.
5. Phase 5, recovery and acceptance: Verify this invariant with a saved fixture: An unreachable URL keeps its bookmark and error state; duplicate imports must not discard existing highlights. State the practical limit: citation graph scale.

## Paid-product capabilities outside this build
- citation graph scale
- team libraries and institutional access
- licensed scholarly metadata
- publisher-specific import reliability

## Implementation prompt
Build me a focused AI research assistant workflow for the personal core of Elicit. Requirements:

- Use exactly this stack: Next.js 15 + TypeScript + SQLite + Playwright. Model source URLs, saved items, reading state, annotations, and exports; keep stable source IDs and timestamps.
- Paid product context: Finds, screens, and extracts evidence from research papers. Build only this DIY scope: Build a private AI research assistant workspace that imports user-supplied URLs or files, extracts metadata, supports notes, and searches the local corpus.
- For AI research assistant, accept a user-supplied URL or file and save its original URL, fetch time, and content hash.
- Extract readable content, deduplicate imports, and preserve notes or highlights by stable source position.
- Search the saved collection and export an item with its source metadata.
- Use a local web page with input, progress, review, and export views. Required input or access: browser automation for user-authorized imports; optional OpenAI API key stored in .env. Keep credentials in .env.
- Acceptance: with one labelled sample, show the input, saved intermediate state, and exported result; verify this invariant: An unreachable URL keeps its bookmark and error state; duplicate imports must not discard existing highlights.
- Out of scope: citation graph scale; team libraries and institutional access. 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.