build(deps): bump @earendil-works/pi-ai from 0.84.1 to 0.84.4

[dependabot]

Bumps @earendil-works/pi-ai from 0.84.1 to 0.84.4.

Release notes

Sourced from @​earendil-works/pi-ai's releases.

v0.84.4

New Features

• Terminal capability overrides — Override detected terminal hyperlink, image, and truecolor support. See Capability Overrides.
• Extension UI prompt events — Integrations can distinguish active agent work from time spent waiting for ctx.ui prompts. See Extension UI prompt events.
• RPC queue clearing — Retrieve and clear queued steering and follow-up messages with clear_queue. See RPC clear_queue.
• Fullscreen selection copy controls — Disable automatic selection copying in fullscreen mode and use Ctrl+X to copy the active selection. See UI & Display.
• DeepSeek V4 Flash Vision (experimental) — Use the vision-capable model through the built-in DeepSeek provider. See API Keys.

Added

• Added ui_prompt_start and ui_prompt_end extension events so host integrations can distinguish active agent work from waiting on user-facing ctx.ui prompts (#8355 by @​cristinaponcela).
• Added detectSupportedImageMimeTypeFromFile() to the public library exports (#8600 by @​xl0).
• Added inherited experimental vision-capable deepseek-v4-flash-vision-exp model support.
• Added transcript usage notices for compaction and branch summaries when cache miss notices are enabled.
• Added RPC clear_queue to retrieve and remove queued steering and follow-up messages (#8432).
• Added environment variables and advanced settings for overriding auto-detected terminal hyperlink, image, and truecolor capabilities (#8665).
• Added fullscreenCopyOnSelect to disable automatic fullscreen selection copy; when disabled, Ctrl+X copies the active text selection before falling back to the last assistant message, while /tree still copies the selected message (#7720).

Fixed

• Fixed toggling thinking visibility clearing partial output from running Bash tools (#8611).
• Fixed Windows shell aborts crashing Pi when taskkill.exe is unavailable on PATH (#6596).
• Fixed resumed sessions corrupting the next appended entry when their JSONL file lacks a trailing newline (#8345).
• Fixed extension messages sent with triggerTurn: false while the agent is running being inserted between a tool call and its result, which made providers that validate message order reject the replayed history. They are now appended once the turn's tool results are in (#8537).
• Fixed compaction and branch summaries forcing toolChoice: "none" (#8649, #8638).
• Fixed large tool results crossing the auto-compaction threshold being sent to the provider before compaction. Pi now compacts between tool execution and the next assistant response in the same run, and restores interactive progress when that run resumes (#6879).
• Fixed Google Vertex requests failing with HttpsProxyAgent is not a constructor when the bundled Node.js runtime uses an HTTP(S) proxy (#8610).
• Fixed saving a default model from a non-empty model scope so it remains available in that scope.
• Fixed inherited @ file autocomplete ranking to prefer direct and shallower matches over similarly ranked nested paths (#8669).
• Fixed inherited OpenAI-compatible streams serializing thinking signatures repeatedly during streaming (#8671).
• Fixed inherited main-screen rendering crashing when image-heavy output exceeded V8's string length limit (#8028).
• Fixed inherited fullscreen double-click word selection splitting paths and kebab-case tokens on / and - (#8676).
• Fixed inherited Cloudflare AI Gateway catalogs omitting supported workers-ai/* passthrough models.
• Fixed inherited OpenAI-compatible reasoning replay to merge consecutive streamed text and summary reasoning_details deltas.
• Fixed inherited OpenRouter reasoning controls so reasoning-mandatory models do not receive effort: "none" (#8614 by @​davidbrai).
• Fixed inherited OpenAI-compatible Chat Completions ignoring an explicitly requested toolChoice when no tools are defined.
• Fixed inherited fragmented Mistral tool calls splitting when continuation chunks omit the tool-call ID (#8387).

v0.84.3

New Features

• PowerShell tool — Use optional native PowerShell command execution on Windows. See PowerShell Tool.
• Safer managed updates — Stage, verify, and atomically activate updates for installer-managed installations. See Install and Manage.
• Model and thinking controls — Select thinking levels with /thinking, search defaults, keep selections session-scoped, and persist them explicitly with Ctrl+S. See Models and Thinking.

Breaking Changes

• Renamed the inherited GoogleThinkingLevel type to GoogleApiThinkingLevel and added ResolvedGoogleThinkingLevel for normalized adapter levels.

... (truncated)

Changelog

Sourced from @​earendil-works/pi-ai's changelog.

[0.84.4] - 2026-08-28

Added

• Added the experimental vision-capable deepseek-v4-flash-vision-exp model to the DeepSeek catalog.

Fixed

• Fixed OpenAI-compatible Chat Completions ignoring an explicitly requested toolChoice when no tools are defined.
• Fixed thinking signature serialization to run once after the signature is complete (#8671).
• Fixed fragmented Mistral tool calls splitting when continuation chunks omit the tool-call ID (#8387).
• Fixed OpenAI-compatible reasoning replay to merge consecutive streamed text and summary reasoning_details deltas.
• Fixed the Cloudflare AI Gateway catalog to include supported workers-ai/* passthrough models omitted by models.dev.
• Fixed OpenRouter reasoning controls by deriving off support and available effort levels from OpenRouter's model metadata, preventing reasoning-mandatory models from receiving effort: "none" (#8614 by @​davidbrai).

[0.84.3] - 2026-08-24

Breaking Changes

• Renamed GoogleThinkingLevel to GoogleApiThinkingLevel and added ResolvedGoogleThinkingLevel for normalized adapter levels.

Added

• Added provider-neutral toolChoice support to simple stream requests.
• Added automatic Anthropic server-side refusal fallback for supported first-party models, including returned-model usage pricing (#8017).
• Added configurable OpenAI-compatible thinking-token budget fields for vLLM, Qwen/SGLang, and llama.cpp servers (#8275 by @​bnsd55).
• Added China-specific ZAI Coding Plan models, including GLM-4.6V vision support, and API-equivalent usage cost estimates for models with published PAYG prices (#8220).
• Added deepseek-v4-pro-0813 to the Qwen Token Plan Individual catalog (#8194).

Changed

• Changed built-in xAI models to use the Responses API with encrypted reasoning replay and made Grok 4.6 the default xAI model (#8124 by @​Jaaneek).
• Changed the Anthropic, Azure OpenAI, Google Generative AI, Google Vertex, Mistral, OpenAI Chat Completions, and OpenAI Responses adapters to send Pi's default User-Agent unless overridden (#8305).

Fixed

• Fixed OpenAI-compatible Chat Completions reasoning replay to preserve and resend assistant-level reasoning_details (reasoning.text, reasoning.summary, and reasoning.encrypted) verbatim and in order (#7994).
• Fixed Anthropic server-side fallback responses being priced with the requested model instead of the returned fallback model (#8285).
• Fixed GitHub Copilot login triggering model-policy rate limits by limiting policy updates, retrying model discovery once, and honoring server retry delays (#7850).
• Fixed Amazon Bedrock dropping and failing to replay opaque redacted reasoning from non-Anthropic models (#8314 by @​seiji).
• Fixed Z.AI Coding Plan models deriving incomplete reasoning-effort metadata, including missing GLM-5.3 low, high, and max levels (#8336).
• Fixed DeepSeek V4 Flash on OpenCode and OpenCode Go omitting its supported low thinking level (#8181 by @​tianshuang).
• Fixed Azure OpenAI Responses ignoring toolChoice in provider-specific stream requests.
• Fixed Amazon Bedrock after_provider_response/onResponse to forward the raw response headers instead of only the synthesized request id header (#8234).
• Fixed Kimi OpenAI-compatible usage reporting so top-level cached_tokens count as cache reads instead of normal input tokens (#8075).
• Fixed Google Generative AI and Vertex AI custom models ignoring thinkingLevelMap, which dropped extended thinking controls (#8135).
• Fixed Xiaomi model catalog generation retaining shut-down MiMo V2 model names after models.dev marked them deprecated (#8187).
• Fixed OpenRouter reasoning controls by deriving off support and available effort levels from OpenRouter's model metadata, preventing reasoning-mandatory models from receiving effort: "none" (#8454).

[0.84.2] - 2026-08-14

... (truncated)

Commits

• b79e4cc Release v0.84.4
• 5ce4afb fix(ai): refresh generated image model catalog
• bba6be9 docs: audit unreleased changelogs
• 6c87d9a fix(ai): merge indexed Mistral tool call chunks
• 56f3f33 fix(ai): remove retired Fireworks turbo router test
• 938109e chore: changelog entries
• 7aab6c2 fix(ai): serialize thinking signature once (#8671)
• 6b36eb5 fix(coding-agent): remove explicit tool choice from compaction calls
• 587be98 feat(ai): add deepseek-v4-flash-vision-exp
• fe37e9f fix(ai): omit tool_choice without tools closes #8607
• Additional commits viewable in compare view