-
Notifications
You must be signed in to change notification settings - Fork 1
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
[Feature]: Configurable VAKRA adapter presets for the M3 benchmark (configured CUGA, never bare main)
enhancementNew feature or requestNew feature or requestevaluation: vakraVakra (m3) evaluationVakra (m3) evaluationStatus: Open.#201 In cuga-project/cuga-eval;Scheduled dependency audit: pip-audit found vulnerabilities
type: securitySecurity vulnerabilities or safety concernsSecurity vulnerabilities or safety concernsStatus: Open.#200 In cuga-project/cuga-eval;- Status: Open.#198 In cuga-project/cuga-eval;
transformers 5.8.1 on macOS: CVE-2026-9856 (path traversal via save_pretrained), never scanned by CI
type: securitySecurity vulnerabilities or safety concernsSecurity vulnerabilities or safety concernsStatus: Open.#193 In cuga-project/cuga-eval;[Evaluation experiment:] Use Indigo to improve CUGA on Appworld
evaluation: appworldAppworld EvalutionAppworld EvalutionStatus: Open.#184 In cuga-project/cuga-eval;- Status: Open.#177 In cuga-project/cuga-eval;
- Status: Open.#174 In cuga-project/cuga-eval;
- Status: Open.#173 In cuga-project/cuga-eval;
- Status: Open.#169 In cuga-project/cuga-eval;
- Status: Open.#168 In cuga-project/cuga-eval;
[Feature]: AppWorld obstacle-escalation rider via the existing special_instructions block (stronger #706 wording, benchmark-scoped)
enhancementNew feature or requestNew feature or requestevaluation: appworldAppworld EvalutionAppworld Evalutionneeds-triageNewly filed, not yet reviewedNewly filed, not yet reviewedStatus: Open.#167 In cuga-project/cuga-eval;[Bug]: stale
errafter a successful retry marks a passing AppWorld SDK task as failedbugSomething isn't workingSomething isn't workingevaluation: appworldAppworld EvalutionAppworld Evalutionpriority: mediumNormal priorityNormal priorityStatus: Open.#153 In cuga-project/cuga-eval;