Feat/saved jobs fix and progress v2

[IfThingsThenStuff]

replaces #167

I wanted the ability to read out my saved jobs - so, I added it. It will handle multiple pages.

Let me know if this is aligned to what you would like to include. Let me know of any changes you think are needed.

Summary

• Add scrape_saved_jobs to LinkedInExtractor — scrapes the LinkedIn jobs tracker page, extracts job IDs from link hrefs, and paginates through results using numbered page buttons
• Add get_saved_jobs MCP tool with progress reporting via on_progress callback
• Cap total_pages with max_pages for accurate progress percentages
• Use Set for O(1) job ID deduplication in the DOM polling function
• Add navigation delay between page clicks consistent with other scraping methods

[greptile-apps]

Greptile Summary

This PR introduces scrape_saved_jobs on LinkedInExtractor and the corresponding get_saved_jobs MCP tool, scraping LinkedIn's /jobs-tracker/ page with full multi-page support, sliding-window pagination refresh, O(1) ID deduplication, and progress reporting. All three issues from the previous review round (pages_visited/sections_requested keys, stale total_pages, missing bug-report checkbox) are resolved in the 1ccb9f9 fixup commit.

Confidence Score: 5/5

• Safe to merge; the one remaining finding is a manifest entry omission that does not affect runtime behavior.
• All prior P0/P1 concerns are fixed, the implementation is well-tested across single-page, multi-page, timeout, and empty-result paths, and the only outstanding item is the missing get_saved_jobs entry in manifest.json — a P2 documentation gap that doesn't break the tool.
• manifest.json — add the get_saved_jobs entry before shipping as a DXT bundle.

Important Files Changed

Filename	Overview
linkedin_mcp_server/scraping/extractor.py	Adds scrape_saved_jobs with correct sliding-window pagination, O(1) ID dedup via prev_ids set, timeout-safe wait_for_function loop, and refreshing total_pages each iteration; reuses shared _EXTRACT_JOB_IDS_JS / _EXTRACT_MAX_PAGE_JS helpers.
linkedin_mcp_server/tools/job.py	Adds get_saved_jobs MCP tool with max_pages (1-10) parameter, progress via _report callback that caps at 99%, and full auth-error handling consistent with other job tools.
tests/test_scraping.py	Adds 5 tests covering single-page, multi-page, timeout, max_pages cap, and empty-result scenarios; verifies progress callbacks, correct job_ids dedup, and documented return keys only.
manifest.json	All existing tools are listed but get_saved_jobs is absent, making the DXT/MCPB bundle manifest incomplete for tool discovery.
.github/ISSUE_TEMPLATE/bug_report.md	Adds get_saved_jobs checkbox to the tool list, addressing the previous review comment.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A[get_saved_jobs tool called] --> B[report_progress 0%]
    B --> C[scrape_saved_jobs\nextract_page /jobs-tracker/]
    C --> D[evaluate _EXTRACT_JOB_IDS_JS\npage 1 IDs]
    D --> E[evaluate _EXTRACT_MAX_PAGE_JS\ninitial total_pages = min_max_pages]
    E --> F[on_progress page=1 total=total_pages]
    F --> G{page_num in range 2..max_pages}
    G -->|next page| H{button\nPage N exists?}
    H -->|no| Z[build id_summary\nreturn result]
    H -->|yes| I[prev_ids = set all_job_ids\nclick button\nsleep _NAV_DELAY]
    I --> J{wait_for_function\nnew ID in DOM?\ntimeout=15s}
    J -->|timeout| Z
    J -->|new ID found| K[scroll_to_bottom\nevaluate EXTRACT_MAIN_TEXT_JS\nevaluate _EXTRACT_JOB_IDS_JS]
    K --> L{new_ids empty?}
    L -->|yes| Z
    L -->|no| M[extend all_job_ids\nrefresh total_pages]
    M --> N[on_progress page=N total=total_pages]
    N --> G
    Z --> O[report_progress 100%\nreturn url + sections + job_ids]

Loading

Prompt To Fix All With AI

This is a comment left during a code review.
Path: manifest.json
Line: 47-99

Comment:
**`get_saved_jobs` missing from `manifest.json` tools list**

Every other tool (`get_job_details`, `search_jobs`, `get_sidebar_profiles`, etc.) is listed in the `tools` array. Omitting `get_saved_jobs` means DXT/MCPB bundle clients that enumerate tools from the manifest won't surface it. The tool still works at runtime via the MCP protocol, but the manifest entry is the authoritative list for bundle discovery.

```suggestion
    {
      "name": "get_job_details",
      "description": "Retrieve specific job posting details using LinkedIn job IDs"
    },
    {
      "name": "get_saved_jobs",
      "description": "Get saved/bookmarked jobs from the LinkedIn job tracker"
    },
    {
      "name": "search_jobs",
      "description": "Search for jobs with filters like keywords and location"
    },
```

How can I resolve this? If you propose a fix, please make it concise.

_{Reviews (8): Last reviewed commit: "Merge branch 'main' into feat/saved-jobs..." | Re-trigger Greptile}

[IfThingsThenStuff]

@stickerdaniel - I've re-applied my old PR to the latest code. It would be great to get this merged in before it drifts again. Thanks

[IfThingsThenStuff]

@stickerdaniel - anything stopping this getting merged?