Open
Conversation
Adds skills/vana-connect/ — a self-contained, agent-agnostic skill for connecting personal data from web platforms. Includes: - SKILL.md: entry point (find/create/run connectors) - SETUP.md: one-time playwright-runner + Chromium install - CREATE.md: full connector creation workflow with validation loop - RECIPES.md: use cases (profile generation, knowledge base, analytics) - scripts/run-connector.cjs: unified runner (replaces test-connector.cjs and agent-run.cjs — JSON output for agents, --pretty for humans) - scripts/scaffold.cjs: template hydration for new connectors - scripts/validate-connector.cjs: structural + output validation - scripts/generate-schemas.cjs: draft schemas from test output - scripts/register.cjs: checksums + registry entry - templates/: connector script, metadata, schema templates - reference/: PAGE-API and extraction patterns docs Auth pattern: try process.env first, fall back to page.requestInput(). Works with Claude Code, Kimi Code, Gemini, or any agent framework. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Replace test-connector.cjs references with run-connector.cjs in README - Point "Building a new connector" to skills/vana-connect/CREATE.md instead of duplicating instructions - Remove AI writing patterns (filler, em-dash reveals, hedging) - Convert em-dashes to -- throughout Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
DOM-based extraction for Max streaming platform. Extracts continue watching (with progress %), personalized recommendations, and My List. Handles multi-language UI (Polish/English) via keyword matching. Filters out live sports to keep data clean. Strips Unicode bidi markers from titles. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Major upgrade: captures JWT token via network interception on page load, then uses httpFetch to call the Max CMS API directly. This gives us: - Real user profile (name, email) from /users/me - Viewing history with timestamps, position, and completion status - My List from the CMS routes API (13 items vs 20 from DOM) - DOM extraction as automatic fallback if token capture fails Replaces the pure DOM approach (v1.0.0) which could only see the homepage snapshot without real viewing history data. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Limitations
Test plan
/my-stuffpage🤖 Generated with Claude Code