issues Search Results · language:Dune language:Python language:JavaScript is:public language:HTML language:JavaScript
Filter by
47M results
Write Evals for Skills
Overview
This issue is to establish a framework for evaluating our 24 AI skills and implement automated evaluations (evals) that
measure their effectiveness and catch regressions. ...
enhancement
jerseycheese
Befund (Gap G5 — Coverage-Qualität)
Das Eval misst nur Quelltreue (Halluzination: Satz gegen Quelle verankert), nicht Erklärgüte. Eine quelltreue Note kann
inhaltlich wertlos sein (Inhaltsverzeichnis-/Randerwähnung ...
TillQuandel
Befund
Die DB-Spalte calibration_labels.agreement_rate (Dashboard-Anzeige) wird falsch berechnet.
generative/calibration/collect.py:165-167:
agree = round(1 - abs(human_hall - llm_hall), 4)
Das rechnet ...
Context
These are non-blocking LOW-severity items intentionally deferred when merging PR #650 (#649). The feature works and is
fail-open; these are robustness/portability/test-hygiene refinements.
Findings ...
bobmatnyc
After using this addon since a lot of weeks, everything is working fine. No issues in retrieving data etc. I know, you
are working on a better/newer integration, but if this take a while, can you please ...
dseichter
CVE-2026-5422 - Medium Severity Vulnerability
details summary img src= https://whitesource-resources.whitesourcesoftware.com/vulnerability_details.png width=19
height=20 Vulnerable Library - b jupyter_server-1.18.1-py3-none-any.whl ...
Mend: dependency security vulnerability
mend-bolt-for-github[bot]
We built a draft prompt template to test if it s feasible to generate test suite using LLM.
Now that we ve finished testing, we need a more systematic way of building prompt template.
We need to consider ...
process
sei-lzhan
CVE-2026-44496 - High Severity Vulnerability
details summary img src= https://whitesource-resources.whitesourcesoftware.com/vulnerability_details.png width=19
height=20 Vulnerable Library - b axios-0.25.0.tgz ...
Mend: dependency security vulnerability
Summary
.form-input (and .form-dropdown-input) ship a disabled: text-colour rule (disabled:text-gray-400 since v4.16.1) but don
t touch the placeholder. Browsers continue rendering input placeholder=\ ...
MarketDataApp
Description
Add a simple Expense Tracker project that allows users to record, categorize, and monitor their daily expenses.
Proposed Solution
- Allow users to add income and expenses.
- Display ...
type:feature
Akhileswar6
Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip! Restrict your search to the title by using the in:title qualifier.
Learn how you can use GitHub Issues to plan and track your work.
Save views for sprints, backlogs, teams, or releases. Rank, sort, and filter issues to suit the occasion. The possibilities are endless.Learn more about GitHub IssuesProTip! Restrict your search to the title by using the in:title qualifier.