feat: layout-agnostic prices, structured product cards, CLI parity + new MCP tools#66
Merged
Merged
Conversation
…LI parity + new MCP tools
Added:
- Layout-agnostic price extraction (prefix/suffix currency, split DOM lines,
nbsp, CH/EU decimals) with per-price context labels
- Structured per-card extraction: browser_products tool + `products` CLI,
extract_schema container mode
- New MCP tools: tabs, dialog, downloads, console, network, autoscroll;
screenshots as MCP resources (screenshot://{sessionId}/last)
- CLI parity: run/products/extract/snapshot/screenshot/inspect (15 commands)
- FUSE_CAPS tool-group filtering, named auth profile, blockResources,
progress notifications, configurable network buffer, self-healing selectors,
weekly anti-bot benchmark
Fixed:
- Booking currency intermediate navigation blanking the target page
- Probe robustness: resilient settle + re-extraction on empty result
- Tab network capture wired before navigation
- mainText strips filter/nav sidebars without dropping product grids
Tooling: Biome linter in CI. Suite green: 292 unit, 20 integration (real Chromium).
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Makes extraction work on real-world e-commerce/OTA layouts, adds structured per-card extraction, brings the CLI to capability parity with the MCP server, and adds several MCP tools. Validated end-to-end on Digitec and Booking.
Changes
Extraction
CHF\n6.90), nbsp/narrow spaces, CH (1'234.56) and EU (1.234,56) decimals. Per-pricecontextlabel.browser_products(MCP) +products(CLI): structured{title, price, currency, url}per product card;extract_schemacontainer mode.mainTextstrips nav/aside/search/filter sub-trees (no more filter-slider prices) while keeping every product-grid card.New MCP tools & config
tabs,dialog,downloads,console,network,autoscroll; screenshots as resources (screenshot://{sessionId}/last).FUSE_CAPSgroup filtering, named authprofile,blockResources, progress notifications, configurable network buffer (FUSE_NETLOG_MAX), self-healing selectors, weekly anti-bot benchmark.CLI parity (9 → 15 commands)
run(multi-step--steps/--steps-file/stdin),products,extract,snapshot,screenshot,inspect.--helplists all 15.Fixes
Test plan
bun test tests/unit— 292 passbun run test:integration— 20 pass (real Chromium)tsc --noEmit+ Biome lint cleanBreaking changes
None — all additions are backward-compatible;
registerResourcessignature change is internal.