| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| 0.121.4 source code.tar.gz | 2026-04-11 | 178.9 MB | |
| 0.121.4 source code.zip | 2026-04-11 | 181.5 MB | |
| README.md | 2026-04-11 | 22.4 kB | |
| Totals: 3 Items | 360.4 MB | 0 | |
0.121.4 (2026-04-10)
Features
- allow per-test opt-out of defaultTest assertions (5e5959e)
- codex: expand Codex SDK eval controls and docs (#8433) (80c3f7f)
- eval: group serial grading by provider (#8509) (d289602)
- examples: add traced openai agents python sdk example (#8354) (6870717)
- http: support structured multipart requests (#8533) (5bac47c)
- japan fiea plugin (#8316) (f330ab3)
- matchers: expose grading provider metadata in GradingResult (#8330) (03cbac6)
- openai: enable chatgpt login via codex (#8327) (5a9cb96)
- providers: add Gemma 4 provider support (#8454) (b0667ed)
- providers: add missing Anthropic SDK features and fix apiKeyRequired bug (#8351) (3060847)
- providers: expand OpenClaw support (#8589) (93f29ec)
- providers: support Codex local images and harden SDK (929790b)
- providers: support multimodal openai agents input (#8397) (4065844)
- redteam: add coding-agent redteam plugins (85a0cfe)
- redteam: add FDA medical plugins (#8456) (30e4ac3)
- redteam: add next coding-agent plugins (5ab3ba4)
- redteam: add teen safety plugins (#8308) (61aa057)
- redteam: enforce max chars per message (#8428) (9af7b7d)
- redteam: use Codex login for default text graders (#8493) (9a6b61b)
Bug Fixes
- app: clarify attack success rate label (#8386) (d088eac)
- app: clarify attack success rate label (#8387) (7482eff)
- app: keep select-all checkbox visible (#8549) (02f9064)
- assertions: apply weights to named scores in assertion results (#8206) (01da019)
- assertions: handle inverse flag in finish-reason handler (#8556) (7155f28)
- assertions: normalize javascript function assertion results (#8377) (a7fafba)
- assertions: normalize negated script assertions (#8485) (fe62243)
- assertions: preserve nested metric weights (#8558) (f3c173d)
- assertions: resolve search-rubric web providers (#8557) (af751aa)
- assertions: skip simulated-user as implicit grader (#8429) (9bc5fdf)
- auth: honour org flag in login command (#8430) (8bdb15d)
- bedrock: preserve zero-valued inference config (#8272) (1cd397e)
- browser: make stealth plugin install explicit (#8434) (aa72a34)
- cache: cache repeated evals by repeat index (#8480) (fbd59a6)
- cli: enforce exact Node engine range (#8380) (1f8e3b4)
- cli: serialize source-map support initialization (#8401) (d4a9d4a)
- code-scan: avoid npm before env for MCP npx (#8515) (7d2eacd)
- code-scan: retry scanner MCP request timeouts (#8545) (4312797)
- deps: pin claude-agent-sdk lockfile to 0.2.87 (#8400) (cbfb91a)
- deps: update dependency @anthropic-ai/sdk to ^0.82.0 (#8510) (cbc96df)
- deps: update dependency @modelcontextprotocol/sdk to ^1.28.0 (416f3cd)
- deps: update dependency @modelcontextprotocol/sdk to ^1.28.0 (5a8a732)
- deps: update dependency @modelcontextprotocol/sdk to ^1.29.0 (#8501) (8821383)
- deps: update dependency @modelcontextprotocol/sdk to ^1.29.0 (#8502) (84db2aa)
- deps: update dependency @openai/agents to ^0.8.1 (#8371) (3f2234c)
- deps: update dependency @openai/agents to ^0.8.1 (#8372) (aecdb11)
- deps: update dependency @openai/agents to ^0.8.2 (#8500) (2b2ce6c)
- deps: update dependency @opentelemetry/exporter-trace-otlp-http to ^0.214.0 (#8379) (462f1e5)
- deps: update dependency ai to ^6.0.138 (#8403) (855e6b3)
- deps: update dependency ai to ^6.0.138 (#8410) (4aa56cf)
- deps: update dependency openai to ^6.33.0 (#8388) (19e04ce)
- deps: update dependency proxy-agent to v8 (#8503) (52953a6)
- deps: update dependency undici to ^7.24.5 (#8411) (3d8a24d)
- deps: update example dependencies (#8367) (977a656)
- deps: update IBM Cloud SDK Core (#8584) (07aaf97)
- deps: update openai packages (#8368) (e5b842a)
- elevenlabs: Disable implicit retries for POST calls (#8541) (34f819a)
- elevenlabs: preserve explicit zero retries (#8358) (493ef88)
- eval: fix bugs and performance issues in evaluator orchestrator (#8481) (36a2788)
- G-Eval wrong scoring for negative criteria (#8259) (e7bba6e)
- handle generated redteam exports during eval (#8301) (ef3f67f)
- harden object guards in comparison checks (#8408) (6e81cc9)
- image should preserve aspect ratio in results table (#8279) (e96d630)
- mcp: preserve falsy values and bound truncation (#8423) (384249e)
- mcp: stabilize withTimeout timeout handling (#8399) (df7d8a9)
- openai: Avoid invalid realtime zero output tokens (#8543) (9cde9b4)
- openai: preserve agents zero token usage (#8283) (6354f4e)
- openai: preserve assistant temperature zero (#8271) (311c0a4)
- parse rate-limit reset headers (#8341) (985ab28)
- preserve configured plugin config in strategy preview (#8326) (e20458e)
- prevent SSTI in conversation-relevance assertion (#8258) (49cbe41)
- prompts: skip recursive rendering for undefined vars (#8394) (8f14e8d)
- providers: align omitDefaults handling across OpenAI and Azure (#8332) (edb5db6)
- providers: deduplicate token refresh retries (#8421) (b78ea03)
- providers: include azure moderation config in cache key (#8349) (c528c5a)
- providers: preserve explicit zero values in Mistral, AI21, and Azure Assistant config (#8555) (7612368)
- providers: prevent max_tokens leaking via passthrough in multi-provider eval (#8449) (b07ca39)
- providers: support systemInstruction for Vertex Claude models (#8522) (ff1cf68)
- providers: support transitive TypeScript provider imports (#8445) (1bb0f39)
- providers: use apiKeyEnvar in error messages across OpenAI providers (#8180) (7008087)
- providers: use rendered prompt in simulated user (#8363) (1f1fa86)
- redteam: avoid nested buttons in transform dialog (#8520) (50c65b4)
- redteam: clarify remote generation guidance (#8252) (45870ab)
- redteam: fix --env-file arg parsing and test flakiness (#8420) (6a920c5)
- redteam: guard missing cross-session leak metadata (#8238) (9924bae)
- redteam: honour generate concurrency config (#8463) (e626ea9)
- redteam: normalize code editor imports (#8306) (057f9e0)
- redteam: pass full conversation history to crescendo grader (#8239) (01619a8)
- redteam: preserve goat zero max turns (#8294) (3077acf)
- redteam: preserve voice crescendo zero max turns (#8310) (5d38b28)
- redteam: preserve zero prompt injection sample (#8291) (16b3221)
- redteam: preserve zero-valued crescendo config (#8273) (9ec66b5)
- redteam: preserve zero-valued custom config (#8274) (a76058f)
- redteam: preserve zero-valued voice crescendo config (#8270) (7d36044)
- redteam: remove dead stores and preserve metadata (#8407) (28d3383)
- redteam: skip unblocking feature checks when disabled (#8266) (c6bdf39)
- redteam: support policy multi-input generation (#8320) (be14251)
- refresh prompt listings after prompt metadata updates (#8120) (85694e9)
- replace undeclared tiny-invariant imports in app (#8309) (f85704f)
- replicate: cache string responses (#8353) (e96fc94)
- respect review language in example test generation (#8334) (3163583)
- sagemaker: preserve zero maxTokens (#8286) (75603d3)
- util: preserve shared references in safeJsonStringify (#8342) (987e8ad)
- watsonx: preserve zero max_new_tokens (#8288) (b489160)
- webui: clear output link feedback timers (#8559) (aeac87d)