# AGENTS.md — Build guide for NewsBlur

## Project scope
Read a personal feed collection, train transparent preferences for authors/tags/phrases and inspect why an article is emphasized or hidden.

Catalogue verdict: yes. The core loop is small enough for a capable coding agent to produce a useful local version in one sitting. For NewsBlur, self-host a feed reader with training filters, folders, and full-text search. The hard boundary is hosted convenience, native apps, story intelligence, and community features, plus capture polish, sync, and content partnerships.
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- Do not publish, spend money, contact customers, delete source data or run irreversible migrations without the project owner's authorization.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
- Runtime and tools: Python, a bounded HTTP worker, feedparser, SQLite FTS5 and a small local web reader.
- Before starting: A reviewed list of permitted RSS/Atom feeds and OPML examples; a running worker is required for scheduled refresh.

## Stack and architecture
- Python, a bounded HTTP worker, feedparser, SQLite FTS5 and a small local web reader
- Data design: Store Subscription, Item, TrainingRule and MatchExplanation; training preferences are reversible and hidden articles remain reachable through an unfiltered view.
- Setup: A reviewed list of permitted RSS/Atom feeds and OPML examples; a running worker is required for scheduled refresh

## Security and data integrity
- Preserve feed GUIDs with a canonical URL fallback. Use conditional requests, per-host backoff and separate fetch errors from empty feeds; sanitize article HTML and prevent private-network URL fetches.
- Keep simple rules distinct from an opaque recommendation model. Start without social sharing or hosted availability guarantees; preserve OPML and article-state exports.
- Keep secrets outside client bundles and exported projects; document what leaves the device and make retention/deletion controls visible.

## Agent implementation rules
- Project rule — domain: Store Subscription, Item, TrainingRule and MatchExplanation; training preferences are reversible and hidden articles remain reachable through an unfiltered view.
- Project rule — scope and recovery: Keep simple rules distinct from an opaque recommendation model. Start without social sharing or hosted availability guarantees; preserve OPML and article-state exports.
- Project rule — acceptance: Train a disliked phrase then discover it inside a useful article; reveal the exact matching rule, override it and recover the article without re-fetching the feed.
- Project rule — delivery: document real setup commands and permissions; do not claim a build, accuracy level, performance result or security certification that has not been demonstrated.

## Optional agent skills and references
- Recommended skill: [modern-python](https://github.com/trailofbits/skills/blob/main/plugins/modern-python/skills/modern-python/SKILL.md) — structure the Python worker or explicitly optional read-only utility with pinned dependencies, typed boundaries and clear failure handling. Follow the maintainer's installation instructions and match its requirements to the chosen runtime.
- Recommended skill: [web-design-guidelines](https://github.com/vercel-labs/agent-skills/blob/main/skills/web-design-guidelines/SKILL.md) — review keyboard access, focus, validation, error recovery and the readable work/review interface or HTML report. Follow the maintainer's installation instructions and match its requirements to the chosen runtime.

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 working slice using synthetic or explicitly authorized non-sensitive examples: Read a personal feed collection, train transparent preferences for authors/tags/phrases and inspect why an article is emphasized or hidden.
- Share a synthetic example export and the acceptance walkthrough; keep real customer, health, financial and source data private: Train a disliked phrase then discover it inside a useful article; reveal the exact matching rule, override it and recover the article without re-fetching the feed.
- State the limits before asking someone to replace their existing tool: Keep simple rules distinct from an opaque recommendation model. Start without social sharing or hosted availability guarantees; preserve OPML and article-state exports.

## Engineering roadmap
1. Phase 1 — Pin the working slice and create its example input: Read a personal feed collection, train transparent preferences for authors/tags/phrases and inspect why an article is emphasized or hidden. Confirm setup: A reviewed list of permitted RSS/Atom feeds and OPML examples; a running worker is required for scheduled refresh.
2. Phase 2 — Implement persistence and write-time invariants before decorating the UI: Store Subscription, Item, TrainingRule and MatchExplanation; training preferences are reversible and hidden articles remain reachable through an unfiltered view.
3. Phase 3 — Connect the working view to real saved state. Preserve feed GUIDs with a canonical URL fallback. Use conditional requests, per-host backoff and separate fetch errors from empty feeds; sanitize article HTML and prevent private-network URL fetches.
4. Phase 4 — Expose the app-specific limits and recovery path in context: Keep simple rules distinct from an opaque recommendation model. Start without social sharing or hosted availability guarantees; preserve OPML and article-state exports.
5. Phase 5 — Walk through this concrete acceptance case and preserve its exported evidence: Train a disliked phrase then discover it inside a useful article; reveal the exact matching rule, override it and recover the article without re-fetching the feed. Finish the README and backup/restore instructions; report unfinished capabilities explicitly.

## Paid-product capabilities outside this build
- hosted convenience, native apps, story intelligence, and community features
- publisher bypasses
- cross-device mobile capture
- high-quality recommendation graph
- licensed full-text feeds

## Implementation prompt
Build the following focused alternative to NewsBlur. Implement the focused workflow below first; the verdict is not evidence of a completed or production-certified build.

WORKING SLICE
Read a personal feed collection, train transparent preferences for authors/tags/phrases and inspect why an article is emphasized or hidden.

SETUP AND ARCHITECTURE
Use Python, a bounded HTTP worker, feedparser, SQLite FTS5 and a small local web reader. Prerequisites: A reviewed list of permitted RSS/Atom feeds and OPML examples; a running worker is required for scheduled refresh. Before integrating anything, record actual versions and permissions, plus model files or provider limits only where used, in the README; make unavailable dependencies visible rather than simulating success.

DOMAIN MODEL AND INVARIANTS
Store Subscription, Item, TrainingRule and MatchExplanation; training preferences are reversible and hidden articles remain reachable through an unfiltered view.

IMPLEMENTATION CONTRACT
Preserve feed GUIDs with a canonical URL fallback. Use conditional requests, per-host backoff and separate fetch errors from empty feeds; sanitize article HTML and prevent private-network URL fetches. Provide an input/setup view, the main work view, and a review/export view appropriate to this workflow. Preserve the last saved state if a job or save fails. Include empty, loading, permission-denied, partial and retryable-error states. Log identifiers and error categories without secret values or unnecessary private content.

APP-SPECIFIC BOUNDARY AND RECOVERY
Keep simple rules distinct from an opaque recommendation model. Start without social sharing or hosted availability guarantees; preserve OPML and article-state exports.

ACCEPTANCE SCENARIO
Train a disliked phrase then discover it inside a useful article; reveal the exact matching rule, override it and recover the article without re-fetching the feed. Also reopen the app after an interrupted operation, confirm the saved record/export remains inspectable, and document the recovery action. These are implementation acceptance requirements, not a claim that this guide has been tested.

DELIVERY
Deliver a runnable repository with migrations or project-format versioning, a non-sensitive example, environment/permission setup, the exact manual acceptance steps, and a backup/export-and-restore walkthrough. Implement the working slice before optional integrations; list any deferred paid-product capabilities honestly. Do not add capabilities outside the working slice just to resemble the original product.

PROJECT RULES FOR AGENTS.md
Keep the domain invariants above executable at the write boundary. Propose scope changes before adding providers or permissions. Never fabricate source evidence, publish results, identity matches or successful delivery. Preserve user originals and require an explicit confirmation for destructive changes or external publication.

## 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.