From abac5d6f29958f94a1cad78367dfd15dcbb48b28 Mon Sep 17 00:00:00 2001 From: Mario Zechner Date: Mon, 23 Mar 2026 02:48:43 +0100 Subject: [PATCH] docs: complete changelog entries for unreleased changes --- packages/ai/CHANGELOG.md | 3 ++- packages/coding-agent/CHANGELOG.md | 13 +++++++++++++ packages/tui/CHANGELOG.md | 1 + 3 files changed, 16 insertions(+), 1 deletion(-) diff --git a/packages/ai/CHANGELOG.md b/packages/ai/CHANGELOG.md index 5f05282c..4c24be87 100644 --- a/packages/ai/CHANGELOG.md +++ b/packages/ai/CHANGELOG.md @@ -4,7 +4,7 @@ ### Added -- Added `requestMetadata` option to `BedrockOptions` for AWS cost allocation tagging; key-value pairs are forwarded to the Bedrock Converse API `requestMetadata` field and appear in AWS Cost Explorer split cost allocation data. +- Added `requestMetadata` option to `BedrockOptions` for AWS cost allocation tagging; key-value pairs are forwarded to the Bedrock Converse API `requestMetadata` field and appear in AWS Cost Explorer split cost allocation data ([#2511](https://github.com/badlogic/pi-mono/pull/2511) by [@wjonaskr](https://github.com/wjonaskr)) - Exported `BedrockOptions` type from the package root entry point, consistent with other provider option types. ### Fixed @@ -12,6 +12,7 @@ - Fixed OpenAI Responses replay for foreign tool-call item IDs by hashing foreign `function_call.id` values into bounded `fc_` IDs instead of preserving backend-specific normalized shapes that OpenAI Codex rejects. - Fixed Anthropic thinking disable handling to send `thinking: { type: "disabled" }` for reasoning-capable models when thinking is explicitly off, and added payload and env-gated end-to-end coverage for the Anthropic provider ([#2022](https://github.com/badlogic/pi-mono/issues/2022)) - Fixed explicit thinking disable handling across Google, Google Vertex, Gemini CLI, OpenAI Responses, Azure OpenAI Responses, and OpenRouter-backed OpenAI-compatible completions. Gemini 3 models now fall back to the lowest supported thinking level when full disable is not supported, and OpenAI/OpenRouter reasoning models now send explicit `none` effort instead of relying on provider defaults ([#2490](https://github.com/badlogic/pi-mono/issues/2490)) +- Fixed OpenAI-compatible completions streams to ignore null chunks instead of crashing ([#2466](https://github.com/badlogic/pi-mono/pull/2466) by [@Cheng-Zi-Qing](https://github.com/Cheng-Zi-Qing)) ## [0.61.1] - 2026-03-20 diff --git a/packages/coding-agent/CHANGELOG.md b/packages/coding-agent/CHANGELOG.md index c3b9dfa3..be5432b9 100644 --- a/packages/coding-agent/CHANGELOG.md +++ b/packages/coding-agent/CHANGELOG.md @@ -2,6 +2,12 @@ ## [Unreleased] +### New Features + +- Built-in tools as extensible ToolDefinitions. Extension authors can now override rendering of built-in read/write/edit/bash/grep/find/ls tools with custom `renderCall`/`renderResult` components. See [docs/extensions.md](docs/extensions.md). +- Unified source provenance via `sourceInfo`. All resources, commands, tools, skills, and prompt templates now carry structured `sourceInfo` with path, scope, and source metadata. Visible in autocomplete, RPC discovery, and SDK introspection. See [docs/extensions.md](docs/extensions.md). +- AWS Bedrock cost allocation tagging. New `requestMetadata` option on `BedrockOptions` forwards key-value pairs to the Bedrock Converse API for AWS Cost Explorer split cost allocation. + ### Breaking Changes - Changed `ToolDefinition.renderCall` and `renderResult` semantics. Fallback rendering now happens only when a renderer is not defined for that slot. If `renderCall` or `renderResult` is defined, it must return a `Component`. @@ -40,6 +46,13 @@ Examples: - Fixed auto-resized image handling to enforce the inline image size limit on the final base64 payload, return text-only fallbacks when resizing cannot produce a safe image, and avoid falling back to the original image in `read` and `@file` auto-resize paths ([#2055](https://github.com/badlogic/pi-mono/issues/2055)) - Fixed `pi update` for git packages to skip destructive reset, clean, and reinstall steps when the fetched target already matches the local checkout ([#2503](https://github.com/badlogic/pi-mono/issues/2503)) - Fixed print and JSON mode to take over stdout during non-interactive startup, keeping package-manager and other incidental chatter off protocol/output stdout ([#2482](https://github.com/badlogic/pi-mono/issues/2482)) +- Fixed cli-highlight auto-detection for languageless code blocks that misidentified prose as programming languages and colored random English words as keywords +- Fixed Anthropic thinking disable handling to send `thinking: { type: "disabled" }` for reasoning-capable models when thinking is explicitly off ([#2022](https://github.com/badlogic/pi-mono/issues/2022)) +- Fixed explicit thinking disable handling across Google, Google Vertex, Gemini CLI, OpenAI Responses, Azure OpenAI Responses, and OpenRouter-backed OpenAI-compatible completions ([#2490](https://github.com/badlogic/pi-mono/issues/2490)) +- Fixed OpenAI Responses replay for foreign tool-call item IDs by hashing foreign IDs into bounded `fc_` IDs +- Fixed OpenAI-compatible completions streams to ignore null chunks instead of crashing ([#2466](https://github.com/badlogic/pi-mono/pull/2466) by [@Cheng-Zi-Qing](https://github.com/Cheng-Zi-Qing)) +- Fixed `truncateToWidth()` performance for very large strings by streaming truncation ([#2447](https://github.com/badlogic/pi-mono/issues/2447)) +- Fixed markdown heading styling being lost after inline code spans within headings ## [0.61.1] - 2026-03-20 diff --git a/packages/tui/CHANGELOG.md b/packages/tui/CHANGELOG.md index 2bfb9ac1..f0a4a2c1 100644 --- a/packages/tui/CHANGELOG.md +++ b/packages/tui/CHANGELOG.md @@ -5,6 +5,7 @@ ### Fixed - Fixed `truncateToWidth()` to stream truncation for very large strings, keep contiguous prefixes, and always terminate truncated SGR styling safely ([#2447](https://github.com/badlogic/pi-mono/issues/2447)) +- Fixed markdown heading styling being lost after inline code spans within headings ## [0.61.1] - 2026-03-20