Repository Issue Activity (beta)

#845 Cookbook idea: a portable-eval-format example alongside evals/ and tool_evaluation/

#830 Memory store never mounts in deployment-created sessions — CMA memory cookbook pattern breaks under scheduled deployments

#839 [BUG] 21 notebooks use claude-opus-4-1, which now returns 404

#838 agents.md

#837 orchestrator_workers.ipynb: context dict is silently dropped from prompts

#708 Security: npmrc files without gitignore

#826 [PROPOSAL] Evaluate Traditional Chinese content with structured outputs

#701 [PROPOSAL] Human-in-the-loop approval for irreversible agent actions

#682 Proposal: guide on citation-faithfulness evals — catching hallucinated/unsupported citations in grounded generation

#713 [PROPOSAL] Cookbook: Adversarial self-verification to cut false positives in agent output

#720 [PROPOSAL] Cookbook: Reconciling an agent's outputs across contexts before they ship

#721 [PROPOSAL] Cookbook: Pipeline vs barrier — composing sub-agents without wasting parallelism

#753 Feature Request: Add scheduled full-repo sweep to lint-format.yml (similar to links.yml)

#680 Sharing: 4 production MCP servers with separated 'Agent Pricing' tier — reproducible pattern

#687 Cookbook Proposal: Claude Agent with Autonomous x402 Micropayments on Solana

#798 Recipe: turn a bad prompt into a great one — automatic prompt scoring + repair with Claude

#797 Related eval: REFUTE scientific critique + calibration benchmark

#486 Add Clarvia integration — MCP tool quality scoring in the agent workflow

#719 [PROPOSAL] Cookbook: Deterministic guardrails as pre/post-action checks on agent tool calls

#779 Cookbook idea: research agents with BGPT scientific evidence MCP

#777 Seasonal Specials Menu - Review Request

#776 Please help

#1 AI Safety Concerns

#761 [BUG] Evaluation scripts still use assistant message prefill and fail on claude-sonnet-4-6

#763 [BUG] sampling_past_max_tokens notebook fails: continuation via assistant prefill is rejected by claude-sonnet-4-6