fix: extension checker API call #218
Merged
Conversation
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
Instead of passing a simple message as the diff payload, the scheduled extension checker now uses lynx to fetch the actual content from both the main conference website and the CFP link page. This provides Claude with real website content to analyze for deadline extensions rather than just receiving an internal trigger message.
The content_contains_cfp_info() function now properly distinguishes between main CFP pages (talks/papers/proposals) and other "Call for X" pages like sponsors, volunteers, reviewers, specialist tracks, etc. Added: - Negative indicators for non-CFP pages (sponsors, volunteers, etc.) - Positive indicators for main CFP pages (call for papers, submit a talk, etc.) - Logic to reject pages where non-CFP indicators dominate without any main CFP indicators present This prevents cfp_link from being incorrectly populated with links to sponsor pages, volunteer sign-ups, or sub-track CFPs.
Instead of using brittle rule-based validation for CFP links in deterministic mode, cfp_link extraction is now handled by Claude in full mode where it can properly understand context. Changes: - Remove content_contains_cfp_info() function (rule-based validation) - Remove cfp_link extraction from run_deterministic_extraction() - Add cfp_link to fields extracted by Claude in build_enrichment_prompt() - Update prompt to instruct Claude to validate cfp_link is main CFP page This ensures cfp_link is only added when AI can confirm it's the main CFP submission page, not sponsors/volunteers/specialist tracks/etc.
Instead of asking Claude to extract cfp_link URLs (which it can't reliably do from content), we now: 1. Find potential cfp_link URLs via pattern matching (deterministic) 2. Fetch content from those URLs 3. Send combined content to Claude for CFP deadline extraction 4. If Claude finds a valid CFP deadline, the cfp_link is validated This means cfp_link is only added when: - We found a potential CFP URL via pattern matching - We successfully fetched its content - Claude confirmed there's a main CFP in that content The cfp_link tracks which URL was fetched so we can add it after Claude validates the content contains CFP info.
The multi-line string with === headers was confusing the YAML parser. Using printf avoids the YAML parsing issue while producing the same output.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
No description provided.