Bump the python group across 1 directory with 6 updates

[dependabot]

Bumps the python group with 6 updates in the / directory:

Package	From	To
python-dotenv	1.2.3	1.2.4
openai	3.16.2	3.24.0
inspect-ai	0.3.265	0.3.276
inspect-evals	0.21.0	0.23.0
ruff	0.16.8	0.16.10
watchfiles	1.2.0	1.3.0

Updates python-dotenv from 1.2.3 to 1.2.4

Release notes

Sourced from python-dotenv's releases.

v1.2.4

Fixed

• dotenv get no longer exits with code 1 for empty string values (KEY=) by [@​ShamikOfficial] in #700
• An unquoted empty value followed by an inline comment (e.g. KEY= # comment) is now parsed as an empty string instead of the comment text by [@​Noethix55555] in #663
• dotenv run --no-override now expands variable references with the same precedence as load_dotenv(override=False), so a value like ${BASE}/suffix uses the existing BASE from the environment instead of the one from the .env file by [@​ROTl24] in #698

Changelog

Sourced from python-dotenv's changelog.

[1.2.4] - 2026-10-01

Fixed

• dotenv get no longer exits with code 1 for empty string values (KEY=) by [@​ShamikOfficial] in #700
• An unquoted empty value followed by an inline comment (e.g. KEY= # comment) is now parsed as an empty string instead of the comment text by [@​Noethix55555] in #663
• dotenv run --no-override now expands variable references with the same precedence as load_dotenv(override=False), so a value like ${BASE}/suffix uses the existing BASE from the environment instead of the one from the .env file by [@​ROTl24] in #698

Commits

• a565c2c Bump version: 1.2.3 → 1.2.4
• 4a7abd0 docs: add 1.2.4 release notes (#663, #698, #700)
• f215c02 fix: dotenv get exits 0 for empty string values (#700)
• 58f2d7c test: make test_run_with_command_flags portable and meaningful (#709)
• e0310e5 fix: honor --no-override when expanding variables in dotenv run (#698)
• a00cb2e docs: add CHANGELOG entry for #663 (fix #600)
• f5485a6 fix: parse empty unquoted value with inline comment as empty string
• See full diff in compare view

Updates openai from 3.16.2 to 3.24.0

Release notes

Sourced from openai's releases.

v3.24.0

3.24.0 (2026-10-02)

Features

• api: add custom voice creation and agent session events (#4013) (e5de2e5)

Bug Fixes

• api: prioritize Python request routing fields (#4014) (ffcc4c4)
• api: prioritize routing fields in parse helpers (#4016) (ba53a10)

v3.23.0

3.23.0 (2026-10-01)

Features

• agents: [1/n] return typed answers from session streams (#4007) (10f8816)
• agents: stage files and download turn artifacts (#4009) (4fc2438)
• api: Add session traces and Realtime translations (#4001) (138e3d1)
• beta: expose typed application actions as agent tools (#4006) (f219c55)
• collect final output from beta Agents streams (157ac4c)

Bug Fixes

• api: allow original image detail in Chat Completions (#4005) (8a136c2)
• api: correct the eval run cancellation endpoint (#4003) (28c5c9a)
• api: retain WebSocket endpoint paths and query parameters (#3999) (7f203fd)
• keep Castiron budget results valid when main advances (#4012) (73f189b)
• responses: avoid replaying uncertain typed sends on reconnect (#4011) (1dbbf61)
• responses: respect send queue limits during reconnect (#4000) (50f95ac)
• use monotonic clock for file processing timeout (#3748) (58aca1d)

Chores

• api: retain WebRTC Live session transport types (#4002) (5c9ace9)
• deps-dev: bump pyright from 1.1.413 to 1.1.414 (#3902) (a91d779)
• deps: bump astral-sh/setup-uv from 10.0.1 to 10.1.0 (#3976) (f15f43c)
• deps: bump CodeQL actions to 4.38.1 (#3974) (fb70d66)

v3.22.1

3.22.1 (2026-09-30)

Bug Fixes

... (truncated)

Changelog

Sourced from openai's changelog.

3.24.0 (2026-10-02)

Features

• api: add custom voice creation and agent session events (#4013) (e5de2e5)

Bug Fixes

• api: prioritize Python request routing fields (#4014) (ffcc4c4)
• api: prioritize routing fields in parse helpers (#4016) (ba53a10)

3.23.0 (2026-10-01)

Features

• agents: [1/n] return typed answers from session streams (#4007) (10f8816)
• agents: stage files and download turn artifacts (#4009) (4fc2438)
• api: Add session traces and Realtime translations (#4001) (138e3d1)
• beta: expose typed application actions as agent tools (#4006) (f219c55)
• collect final output from beta Agents streams (157ac4c)

Bug Fixes

• api: allow original image detail in Chat Completions (#4005) (8a136c2)
• api: correct the eval run cancellation endpoint (#4003) (28c5c9a)
• api: retain WebSocket endpoint paths and query parameters (#3999) (7f203fd)
• keep Castiron budget results valid when main advances (#4012) (73f189b)
• responses: avoid replaying uncertain typed sends on reconnect (#4011) (1dbbf61)
• responses: respect send queue limits during reconnect (#4000) (50f95ac)
• use monotonic clock for file processing timeout (#3748) (58aca1d)

Chores

• api: retain WebRTC Live session transport types (#4002) (5c9ace9)
• deps-dev: bump pyright from 1.1.413 to 1.1.414 (#3902) (a91d779)
• deps: bump astral-sh/setup-uv from 10.0.1 to 10.1.0 (#3976) (f15f43c)
• deps: bump CodeQL actions to 4.38.1 (#3974) (fb70d66)

3.22.1 (2026-09-30)

Bug Fixes

• api: correct the missing authentication error message (#3993) (51692ed), closes #3962
• transform NotRequired typed dictionary fields (#3995) (5544c18)

... (truncated)

Commits

• 637f1b8 release: 3.24.0 (#4015)
• e5de2e5 feat(api): add custom voice creation and agent session events (#4013)
• ba53a10 fix(api): prioritize routing fields in parse helpers (#4016)
• ffcc4c4 fix(api): prioritize Python request routing fields (#4014)
• 93bd2aa release: 3.23.0 (#3996)
• 73f189b fix: keep Castiron budget results valid when main advances (#4012)
• 1dbbf61 fix(responses): avoid replaying uncertain typed sends on reconnect (#4011)
• 4fc2438 feat(agents): stage files and download turn artifacts (#4009)
• a91d779 chore(deps-dev): bump pyright from 1.1.413 to 1.1.414 (#3902)
• fb70d66 chore(deps): bump CodeQL actions to 4.38.1 (#3974)
• Additional commits viewable in compare view

Updates inspect-ai from 0.3.265 to 0.3.276

Changelog

Sourced from inspect-ai's changelog.

0.3.276 (02 October 2026)

• LiteLLM Proxy: Gemini tool results that are JSON objects are sent under content, as the google provider sends them; LiteLLM otherwise passed the object as the function response itself and Vertex rejected documents with $ref keys (an OpenAPI spec read with curl) with a 400.

0.3.275 (01 October 2026)

• LiteLLM Proxy: Gemini models accept any reasoning_effort, mapped to the levels or thinking budgets the native Google provider uses.
• LiteLLM Proxy: Gemini tool calls replayed through the proxy keep their thought signatures when the proxy's model_name does not contain "gemini" (LiteLLM otherwise replaced them with a placeholder); requests with tools carry the function-calling hint; and a turn returned as a malformed function call (no text or tool call, or a call written as code) is retried with a corrective exchange, as with the google provider.

0.3.274 (01 October 2026)

• Sandbox tools: the root check now runs once at sample start, before solver/agent execution begins; an inconclusive check warns before falling back to the sandbox's default user, and a check that could not run, or a later root failure, is an error.
• Local sandbox: exec(user=...) now rejects unsupported users instead of ignoring them; the current effective user's name or UID is accepted on POSIX.
• Control Channel: inspect ctl ... --log-dir now shows running and completed-but-unflushed samples, with current counts and their events, for evals run with --log-shared.
• OpenAI Compatible: Fixed concurrent sample failures during credential refresh, including OpenRouter evaluations on Hawk, and added an overridable ModelAPI.refresh_credentials() for model API extensions.
• OpenAI: Biological-risk policy responses now produce content-filter stops instead of failing samples.
• Log viewer: Tasks can set ViewerConfig(trust_content=False) to have the viewer show all of a log's content as plain text, with no markdown, media, or clickable links.
• Log viewer: inspect view --no-trust-content (or INSPECT_VIEW_TRUST_CONTENT=false) shows every log's content as plain text.

0.3.273 (29 September 2026)

• Anthropic: Prompts can mark a stable prefix for caching while allowing the remaining text to change.
• OpenAI: GPT-5.6 and later support explicit prompt-cache boundaries; other models retain automatic caching.
• OpenAI: Support for GPT-6.1 Sol (gpt-6.1-sol), including its context window and output limits.
• Mistral: Support for mistralai 3.0, which is now the minimum required version.
• Logs: Reading samples with resolve_attachments is much faster for long conversations; in full mode, deeply nested model API call content may keep two more nesting levels.
• Bugfix: self_critique(), and model_graded_qa()/model_graded_fact() with model_role=None, now critique or grade with the correct model when one task is evaluated against several models.
• Reading a remote .eval log from a non-S3 filesystem (e.g. gs://, az://) no longer stalls other running work for the whole download.
• S3: Streaming uploads now read each multipart part from the source file in a single worker-thread call rather than 32 chunked calls, reducing event-loop wakeups during log flushes.
• Recording an eval or model error no longer spends seconds syntax highlighting its traceback: the stored ANSI traceback keeps its frames and source snippets, without syntax colouring.
• Bugfix: subprocess() and Docker sandbox exec() no longer intermittently fail with BrokenPipeError when the command exits before reading its input.

0.3.272 (28 September 2026)

• Anthropic: Support for Claude Sonnet 5.5 (claude-sonnet-5-5): reasoning_effort="none" turns off up-front thinking, forced tool choice degrades to auto, and computer use uses the computer toolset on the Claude API and Vertex.
• Tools: web_browser() is deprecated, logs a warning when called, and will be removed in a future release.
• Inspect View: Shard logs under <name>.shards/ are hidden once a successful merged log covers them (inspect view --show-shards to list them); bundles leave them out.
• Computer tool: click actions called without a coordinate now click at the current cursor position instead of failing, and the tool description states which actions require one.
• Computer tool: back_click and forward_click now work (with a rebuilt aisiuk/inspect-computer-tool image); previously they failed inside the container regardless of arguments.

0.3.271 (26 September 2026)

• CLI: New --extra-headers and --extra-body options for inspect eval and inspect eval-set, taking an inline JSON or YAML mapping.
• Google: Token counting falls back to a local estimate, with a warning, when the countTokens endpoint is unavailable or rejects the request.

0.3.270 (25 September 2026)

• eval_set: a task whose dataset GREW since its last run (a strict superset, with stable sample ids) is now topped up in place — the prior samples are reused and only the newly added samples run — instead of re-running the whole dataset. A shrunk dataset, or growth without stable ids, still triggers a full re-run.
• LiteLLM Proxy: GPT-5 and later, o-series and Codex models served by OpenAI use the Responses API by default, streamed, carrying reasoning between turns (-M responses_api=false to opt out).
• LiteLLM Proxy: Key and team aliases resolve to the model they route to, instead of failing for lack of model info.

... (truncated)

Commits

• 93f7182 update changelog for release
• 5f01b5a gemini litellm proxy tool call escaping (#5661)
• c05398d update changelog for release
• 79483c1 malformed tool call handling for gemini on litellm-proxy (#5657)
• 1d2fab1 litellm_proxy: bind gemini reasoning effort (#5656)
• 617d06c update changelog for release
• 6bea9cd Let tasks and the viewer mark log content untrusted so it shows as plain text...
• 0321960 Fix S3 finalizer test failing with aiobotocore 3.9.2 (#5633)
• 674e628 Treat OpenAI biological-risk blocks as content filters (#5627)
• ddb404e Document sandbox directory security review requirements (#5620)
• Additional commits viewable in compare view

Updates inspect-evals from 0.21.0 to 0.23.0

Release notes

Sourced from inspect-evals's releases.

v0.23.0

Existing Evals

• AgentHarm: Harmfulness Potential in AI Agents: Remove an unreachable float-type guard from avg_full_score so it filters samples the same way as the sibling metrics. No effect on reported scores. (@​EastZeus)

• Humanity's Last Exam: Register the inspect_evals/hle/regex_judge and inspect_evals/hle/json_judge scorers when the package is imported, so inspect eval inspect_evals/hle no longer fails with a registry lookup error at task construction.

• MBPP (v4-A): Completion extraction now handles a python3 label, an unclosed fence, a fence after prose on the same line, a closing fence glued to code, and empty or stray fences. These completions used to be scored on the raw text or an empty block. Blocks labelled py and python are now taken in order of position.

Other

• Documentation: The site follows the reader's system light or dark setting, and a navbar toggle overrides it. Colours come from a single brand file with light and dark values, and page styles use the theme's colours instead of fixed ones, so code blocks, lint badges and search highlights read correctly in both.

• The lint results dashboard supports the register-v2 policy and inspect-evals-lint schema version 3, where suppressed checks count as met, and shows how many checks were suppressed beside each count. Existing register-v1 reports remain supported.

• The lint results dashboard supports the security category and inspect-evals-lint schema version 2. Existing reports remain supported.

• extract_code_block in inspect_evals.utils.code takes an optional accept predicate and returns the first block it accepts, falling back to the first block. The new defines(name) predicate accepts a block that parses and defines the required function, so an eval can skip a usage example or a draft that does not parse. The parser fixes above apply to every caller.

• New extract_code_blocks returns every language-labelled block in order, or else every unlabelled one, for evals that take the last block or all blocks. extract_code_block is built on it. The cpp language also matches cc and cxx labels.

• Add shuffle_and_seed to inspect_evals.utils, which splits a shuffle: bool | int task parameter into a loader's shuffle and seed arguments, so a task can default its shuffle to a seed without a separate seed parameter. BEST_PRACTICES.md now recommends it for datasets whose records are grouped.

v0.22.0

Existing Evals

• PaperBench (v5-C): Criteria the judge could not grade are no longer scored as criteria the reproduction did not meet. They are left out of the weighted rubric average on both sides of the division, after bounded retries: judge replies are classified before use, so a refusal, an empty reply, or a file ranking that names nothing readable is retried rather than parsed as an essay or graded against no files, and a grading prompt that overflows the context window is retried once with half the file budget before erroring the sample. A rubric with no gradeable criterion returns Score.unscored(), and failures of the machinery around the judge error the sample. Score.value is now a dict with keys score and graded_weight_fraction, the second reporting how much of the rubric was graded. The score mean can move in either direction: per-sample scores rise when a judge reply was ungradeable, and samples whose grader call raised now leave the denominator.

• AgentBench (v4-B): Add sandbox_type/sandbox_config provider configuration following the cross-eval sandbox-config pattern (#1115).

• PaperBench (v3-C): Add a task-level sandbox_config: SandboxEnvironmentSpec override for non-default providers (#1115).

• Mind2Web-SC (v3-B): Load the full 200-sample dataset instead of truncating to the first 10 samples via hardcoded slice, and disambiguate sample IDs across ALLOW/DENY variants sharing the same task annotation.

• HLE: original_accuracy now follows CAIS's rounded percentage calculation using the selected dataset size after dataset filters and before run limits. The dataset size is recorded in sample metadata, so omitted questions and missing judgments remain in the denominator for static and rolling datasets. This metric requires single-attempt results and dataset-size metadata; other metric calculations are unchanged.

• HLE: the default run config now evaluates the HLE-Verified gold subset (only_hle_verified_gold: true); original.yaml keeps the full CAIS dataset. Default-run scores are not comparable with earlier versions.

• HLE: select metrics and their arguments through the scorer spec's metrics field in run configs. Metric factories resolve through Inspect registration. Score calculations are unchanged, but each shipped config now selects its own metrics: the default run reports hle/accuracy, hle/stderr, hle/cerr and hle/unscored and no longer emits hle/original_accuracy; original.yaml reports hle/original_accuracy, hle/stderr, hle/cerr and hle/unscored. Direct use of the HLE judge factories now requires metric selection on the containing task. Scorer and metric names in the spec are Inspect registry names, and everything HLE defines is registered under an eval-qualified name (inspect_evals/hle/regex_judge, inspect_evals/hle/accuracy, ...), so the score columns are now hle/regex_judge and hle/regex_judge1 (hle/json_judge under original.yaml) and every metric row carries the hle/ prefix.

• PaperBench (v4-B): Update blacklist URL monitor to identify scp-like SSH clones git clone git@github.com:owner/repo.git. Deviates from original paper implementation, however, as the paper implementation outlines that the blacklist monitor is used in addition to human checking, this change is helps with the human check and aligns with the intent of the paper.

• BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions (v4-A), PAWS: Paraphrase Adversaries from Word Scrambling (v4-A), MaCBench: Probing the limitations of multimodal language models for chemistry and materials research (v2-A): A completion that does not match the answer pattern is now scored INCORRECT with reason="invalid_response_format", rather than NOANSWER. This follows pattern()/answer() upstream in inspect_ai#4629, which attributes an unmatched completion to the model under test. Headline metrics are unaffected: value_to_float maps both NOANSWER and INCORRECT to 0.0, so accuracy and stderr are unchanged on every run. What moves is where the non-answer is recorded, from the score value to Score.reason; log analysis that counts NOANSWER values should read reason instead.

• BBH: Challenging BIG-Bench Tasks (v3-A), ChemBench: Are large language models superhuman chemists? (v3-B), ∞Bench: Extending Long Context Evaluation Beyond 100K Tokens (v3-A) and WorldSense: Grounded Reasoning Benchmark (v3-A): The same pattern()/answer() change as above. It reaches BBH's binary-choice subsets, ChemBench's multiple-choice questions, infinite_bench_code_run and all of WorldSense; the other scorers in those evals are unchanged. Headline metrics are unaffected for the same reason.

• KernelBench: Use a standalone uv project to pin sandbox dependencies independently of the host lockfile, so host dependency updates do not require rebuilding the published image.

... (truncated)

Changelog

Sourced from inspect-evals's changelog.

[0.23.0] — 2026-10-01

Existing Evals

• AgentHarm: Harmfulness Potential in AI Agents: Remove an unreachable float-type guard from avg_full_score so it filters samples the same way as the sibling metrics. No effect on reported scores. (@​EastZeus)

• Humanity's Last Exam: Register the inspect_evals/hle/regex_judge and inspect_evals/hle/json_judge scorers when the package is imported, so inspect eval inspect_evals/hle no longer fails with a registry lookup error at task construction.

• MBPP (v4-A): Completion extraction now handles a python3 label, an unclosed fence, a fence after prose on the same line, a closing fence glued to code, and empty or stray fences. These completions used to be scored on the raw text or an empty block. Blocks labelled py and python are now taken in order of position.

Other

• Documentation: The site follows the reader's system light or dark setting, and a navbar toggle overrides it. Colours come from a single brand file with light and dark values, and page styles use the theme's colours instead of fixed ones, so code blocks, lint badges and search highlights read correctly in both.

• The lint results dashboard supports the register-v2 policy and inspect-evals-lint schema version 3, where suppressed checks count as met, and shows how many checks were suppressed beside each count. Existing register-v1 reports remain supported.

• The lint results dashboard supports the security category and inspect-evals-lint schema version 2. Existing reports remain supported.

• extract_code_block in inspect_evals.utils.code takes an optional accept predicate and returns the first block it accepts, falling back to the first block. The new defines(name) predicate accepts a block that parses and defines the required function, so an eval can skip a usage example or a draft that does not parse. The parser fixes above apply to every caller.

• New extract_code_blocks returns every language-labelled block in order, or else every unlabelled one, for evals that take the last block or all blocks. extract_code_block is built on it. The cpp language also matches cc and cxx labels.

• Add shuffle_and_seed to inspect_evals.utils, which splits a shuffle: bool | int task parameter into a loader's shuffle and seed arguments, so a task can default its shuffle to a seed without a separate seed parameter. BEST_PRACTICES.md now recommends it for datasets whose records are grouped.

[0.22.0] — 2026-09-25

Existing Evals

• PaperBench (v5-C): Criteria the judge could not grade are no longer scored as criteria the reproduction did not meet. They are left out of the weighted rubric average on both sides of the division, after bounded retries: judge replies are classified before use, so a refusal, an empty reply, or a file ranking that names nothing readable is retried rather than parsed as an essay or graded against no files, and a grading prompt that overflows the context window is retried once with half the file budget before erroring the sample. A rubric with no gradeable criterion returns Score.unscored(), and failures of the machinery around the judge error the sample. Score.value is now a dict with keys score and graded_weight_fraction, the second reporting how much of the rubric was graded. The score mean can move in either direction: per-sample scores rise when a judge reply was ungradeable, and samples whose grader call raised now leave the denominator.

• AgentBench (v4-B): Add sandbox_type/sandbox_config provider configuration following the cross-eval sandbox-config pattern (#1115).

• PaperBench (v3-C): Add a task-level sandbox_config: SandboxEnvironmentSpec override for non-default providers (#1115).

• Mind2Web-SC (v3-B): Load the full 200-sample dataset instead of truncating to the first 10 samples via hardcoded slice, and disambiguate sample IDs across ALLOW/DENY variants sharing the same task annotation.

• HLE: original_accuracy now follows CAIS's rounded percentage calculation using the selected dataset size after dataset filters and before run limits. The dataset size is recorded in sample metadata, so omitted questions and missing judgments remain in the denominator for static and rolling datasets. This metric requires single-attempt results and dataset-size metadata; other metric calculations are unchanged.

• HLE: the default run config now evaluates the HLE-Verified gold subset (only_hle_verified_gold: true); original.yaml keeps the full CAIS dataset. Default-run scores are not comparable with earlier versions.

• HLE: select metrics and their arguments through the scorer spec's metrics field in run configs. Metric factories resolve through Inspect registration. Score calculations are unchanged, but each shipped config now selects its own metrics: the default run reports hle/accuracy, hle/stderr, hle/cerr and hle/unscored and no longer emits hle/original_accuracy; original.yaml reports hle/original_accuracy, hle/stderr, hle/cerr and hle/unscored. Direct use of the HLE judge factories now requires metric selection on the containing task. Scorer and metric names in the spec are Inspect registry names, and everything HLE defines is registered under an eval-qualified name (inspect_evals/hle/regex_judge, inspect_evals/hle/accuracy, ...), so the score columns are now hle/regex_judge and hle/regex_judge1 (hle/json_judge under original.yaml) and every metric row carries the hle/ prefix.

• PaperBench (v4-B): Update blacklist URL monitor to identify scp-like SSH clones git clone git@github.com:owner/repo.git. Deviates from original paper implementation, however, as the paper implementation outlines that the blacklist monitor is used in addition to human checking, this change is helps with the human check and aligns with the intent of the paper.

• BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions (v4-A), PAWS: Paraphrase Adversaries from Word Scrambling (v4-A), MaCBench: Probing the limitations of multimodal language models for chemistry and materials research (v2-A): A completion that does not match the answer pattern is now scored INCORRECT with reason="invalid_response_format", rather than NOANSWER. This follows pattern()/answer() upstream in inspect_ai#4629, which attributes an unmatched completion to the model under test. Headline metrics are unaffected: value_to_float maps both NOANSWER and INCORRECT to 0.0, so accuracy and stderr are unchanged on every run. What moves is where the non-answer is recorded, from the score value to Score.reason; log analysis that counts NOANSWER values should read reason instead.

• BBH: Challenging BIG-Bench Tasks (v3-A), ChemBench: Are large language models superhuman chemists? (v3-B), ∞Bench: Extending Long Context Evaluation Beyond 100K Tokens (v3-A) and WorldSense: Grounded Reasoning Benchmark (v3-A): The same pattern()/answer() change as above. It reaches BBH's binary-choice subsets, ChemBench's multiple-choice questions, infinite_bench_code_run and all of WorldSense; the other scorers in those evals are unchanged. Headline metrics are unaffected for the same reason.

• KernelBench: Use a standalone uv project to pin sandbox dependencies independently of the host lockfile, so host dependency updates do not require rebuilding the published image.

... (truncated)

Commits

• bd59dd3 Prepare release v0.23.0 (#2612)
• d291bbf Update PR process: require authors be on the contributors list or be assigned...
• 190dfa2 feat(utils): accept predicate, extract_code_blocks and parser fixes for extra...
• 6c97878 fix(agentharm): include integer scores in avg_full_score (#2154)
• 0be63e5 feat(utils): shuffle_and_seed for seeded-by-default shuffle parameters (#2595)
• d31de3b docs(pr-template): say what the Related issue section must contain (#2596)
• 12d04e0 Lint dashboard: support the register-v2 policy and lint schema 3 (#2580)
• 95036ea Accept lint schema version 2 in the internal lint document test (#2578)
• e645ed2 Use inspect-evals-lint 0.9.0 (#2576)
• 80d5436 Support security checks in the lint results dashboard (#2570)
• Additional commits viewable in compare view

Updates ruff from 0.16.8 to 0.16.10

Release notes

Sourced from ruff's releases.

0.16.10

Release Notes

Released on 2026-10-01.

Preview features

• Add a migration guide for categories (#28087)
• [pyupgrade] Add rule for context manager iterator annotations (UP052) (#29000)

Performance

• Reduce memory used by diagnostics (#28951)

Server

• Avoid running uv format in untrusted workspaces (#28873)

Documentation

• Fix links to moved changelog sections and renamed mdtests (#28941)
• Add Python 3.15 as a supported version (#28907)
• Add ty as a type checker example (#28906)

Other changes

• Update Rust toolchain to 1.99 and MSRV to 1.97 (#29047)

Contributors

• @​ntBre
• @​spaceone
• @​HardMax71
• @​charliermarsh

Install ruff 0.16.10

Install prebuilt binaries via shell script

curl --proto '=https' --tlsv1.2 -LsSf https://releases.astral.sh/github/ruff/releases/download/0.16.10/ruff-installer.sh | sh

Install prebuilt binaries via powershell script

powershell -ExecutionPolicy Bypass -c "irm https://releases.astral.sh/github/ruff/releases/download/0.16.10/ruff-installer.ps1 | iex"

Download ruff 0.16.10

... (truncated)

Changelog

Sourced from ruff's changelog.

0.16.10

Released on 2026-10-01.

Preview features

• Add a migration guide for categories (#28087)
• [pyupgrade] Add rule for context manager iterator annotations (UP052) (#29000)

Performance

• Reduce memory used by diagnostics (#28951)

Server

• Avoid running uv format in untrusted workspaces (#28873)

Documentation

• Fix links to moved changelog sections and renamed mdtests (#28941)
• Add Python 3.15 as a supported version (#28907)
• Add ty as a type checker example (#28906)

Other changes

• Update Rust toolchain to 1.99 and MSRV to 1.97 (#29047)

Contributors

• @​ntBre
• @​spaceone
• @​HardMax71
• @​charliermarsh

0.16.9

Released on 2026-09-24.

Preview features

• [ruff] Avoid false positives for overloaded division (RUF069) (#28309)

Bug fixes

• [flake8-bugbear] Avoid false positives for calls with keyword arguments (B009, B010, B043) (#28776)
• [flake8-tidy-imports] Allow lazy imports to be used in deferred annotations (TID255) (#28767)

Rule changes

• Update LibCST-based fixes for Python 3.15 (#28616)

... (truncated)

Commits

• 3265ed1 Bump version to 0.16.10 (#29055)
• e786964 Authorize shared PR security-review workflow to publish findings (#29052)
• a81291e [ty] Defer uv workspace discovery until after project configuration (#28525)
• b6a74d2 [ty] Refresh uv project metadata when uv files change (#28529)
• 41d30df Update Rust toolchain to 1.99 and MSRV to 1.97 (#29047)
• 317e0a3 [ty] Bound nested callable signature display (#29049)
• 8546752 [ty] Fix member lookup on union-bounded type variables (#29018)
• 56180bc [ty] Specialize instance members once (#29043)
• aa9a1ff [ty] Avoid stale I/O diagnostics when closing deleted files (#28988)
• 2d25346 [ty] Improve unresolved-import documentation (#29039)
• Additional commits viewable in compare view

Updates watchfiles from 1.2.0 to 1.3.0

Release notes

Sourced from watchfiles's releases.

v1.3.0 2026-09-21

What's Changed

• Bump pyo3 from 0.28.3 to 0.29.0 by @​dependabot[bot] in samuelcolvin/watchfiles#382
• Add Python 3.14t to Windows wheel builds by @​1998LJ in samuelcolvin/watchfiles#393
• Build abi3 wheels for CPython by @​samuelcolvin in samuelcolvin/watchfiles#396
• prep v1.3.0 release by @​samuelcolvin in samuelcolvin/watchfiles#397

New Contributors

• @​1998LJ made their first contribution in samuelcolvin/watchfiles#393

Full Changelog: samuelcolvin/watchfiles@v1.2.0...v1.3.0

Commits

• e74b8fc prep v1.3.0 release (#397)
• 026c971 Build abi3 wheels for CPython (#396)
• 59099d2 Add Python 3.14t to Windows wheel builds (#393)
• 42257e8 Bump pyo3 from 0.28.3 to 0.29.0 (#382)
• See full diff in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore <dependency name> major version will close this group update PR and stop Dependabot creating any more for the specific dependency's major version (unless you unignore this specific dependency's major version or upgrade to it yourself)
• @dependabot ignore <dependency name> minor version will close this group update PR and stop Dependabot creating any more for the specific dependency's minor version (unless you unignore this specific dependency's minor version or upgrade to it yourself)
• @dependabot ignore <dependency name> will close this group update PR and stop Dependabot creating any more for the specific dependency (unless you unignore this specific dependency or upgrade to it yourself)
• @dependabot unignore <dependency name> will remove all of the ignore conditions of the specified dependency
• @dependabot unignore <dependency name> <ignore condition> will remove the ignore condition of the specified dependency and ignore conditions