-
Notifications
You must be signed in to change notification settings - Fork 0
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
Experiment with recurring research, structured hypotheses, and competing multi-step plans
ready-for-agentReady for an implementation agentReady for an implementation agentStatus: Open.#204 In justsml/ExploitHunter.app;RL PoC Phase 3: Run matched specialization and decomposition ablations
difficulty: XLVery large design and implementation scopeVery large design and implementation scopeenhancementNew feature or requestNew feature or requestpriority: highHigh-value safety or evaluation integrity workHigh-value safety or evaluation integrity workready-for-humanRequires maintainer judgment, external access, paid execution, or human reviewRequires maintainer judgment, external access, paid execution, or human reviewStatus: Open.#203 In justsml/ExploitHunter.app;Deepen canonical Benchmark Run publication and reprojection
difficulty: LLarge cross-module implementation scopeLarge cross-module implementation scopeenhancementNew feature or requestNew feature or requestpriority: highHigh-value safety or evaluation integrity workHigh-value safety or evaluation integrity workready-for-agentReady for an implementation agentReady for an implementation agentStatus: Open.#195 In justsml/ExploitHunter.app;Make the resolved Mastra model runtime the sole invocation seam
difficulty: LLarge cross-module implementation scopeLarge cross-module implementation scopeenhancementNew feature or requestNew feature or requestpriority: highHigh-value safety or evaluation integrity workHigh-value safety or evaluation integrity workready-for-agentReady for an implementation agentReady for an implementation agentStatus: Open.#194 In justsml/ExploitHunter.app;Give Benchmark Run attempt lifecycle one durable owner
difficulty: LLarge cross-module implementation scopeLarge cross-module implementation scopeenhancementNew feature or requestNew feature or requestpriority: highHigh-value safety or evaluation integrity workHigh-value safety or evaluation integrity workready-for-agentReady for an implementation agentReady for an implementation agentStatus: Open.#193 In justsml/ExploitHunter.app;Deepen the Browser Benchmark Run lifecycle
difficulty: XLVery large design and implementation scopeVery large design and implementation scopeenhancementNew feature or requestNew feature or requestpriority: highHigh-value safety or evaluation integrity workHigh-value safety or evaluation integrity workready-for-agentReady for an implementation agentReady for an implementation agentStatus: Open.#192 In justsml/ExploitHunter.app;Decide the OSS-to-device transition and first device campaign
wayfinder:grillingWayfinder HITL design decisionWayfinder HITL design decisionStatus: Open.#136 In justsml/ExploitHunter.app;Decide the first executed OSS campaign and its exit criteria
wayfinder:grillingWayfinder HITL design decisionWayfinder HITL design decisionStatus: Open.#135 In justsml/ExploitHunter.app;Decide device target admission, recovery, and non-mutation contract
wayfinder:grillingWayfinder HITL design decisionWayfinder HITL design decisionStatus: Open.#134 In justsml/ExploitHunter.app;Decide the AgentController skill taxonomy and promotion pipeline
wayfinder:grillingWayfinder HITL design decisionWayfinder HITL design decisionStatus: Open.#133 In justsml/ExploitHunter.app;Decide harness-study admission and attribution design
wayfinder:grillingWayfinder HITL design decisionWayfinder HITL design decisionStatus: Open.#132 In justsml/ExploitHunter.app;Decide behavioral-anomaly ranking and impact-validation methodology
wayfinder:grillingWayfinder HITL design decisionWayfinder HITL design decisionStatus: Open.#131 In justsml/ExploitHunter.app;