* agent host init
* Agent host: Copilot SDK integration with chat UI
* Agent host: direct MessagePort, logging, SDK wrapper, env fix
* Refactoring and cleanup
* Copilot-authored message: Agent-host tool rendering, protocol, and session fixes
Tool invocation rendering:
- Emit tool_start/tool_complete as ChatToolInvocation (not progressMessage)
- Shell tools (bash/powershell) render as terminal command blocks with
IChatTerminalToolInvocationData, output, and exit codes
- Non-shell tools render via invocationMessage/pastTenseMessage (markdown)
- Filter out report_intent (hidden internal tool)
Agent-agnostic protocol:
- IPC events carry display-ready fields (displayName, invocationMessage,
pastTenseMessage, toolInput, toolOutput, toolKind, language)
- All Copilot CLI-specific logic in copilotToolDisplay.ts with typed
interfaces for known tools (CopilotToolName enum, parameter types)
- Renderer never references specific SDK tool names
Session fixes:
- Resumed sessions show tool invocations in history (getSessionMessages
now returns tool events alongside messages)
- Fixed 'already has a pending request' on resumed sessions by
conditionally providing interruptActiveResponseCallback
- Fixed event filtering for resumed sessions (sessionId override in
_trackSession)
Documentation:
- Split parity.md into design.md (decisions) and backlog.md (tasks)
- Updated architecture.md, sessions.md with cross-references
- Added maintenance notes to all docs
* Copilot-authored message: Model picker, session class, DI and test cleanup
* Cleanups
* stuff
* add diagram
* Add claude agent
* Clean up
* Copy some build script changes from #295817
* Simplify
* Update docs
* Register agent-host via chatSessions contribution API, reduce peripheral diff
* Cleanup
* Don't ship stuff in stable
* Dynamic agent discovery via listAgents() IPC
Replace hardcoded per-provider contributions with a single
AgentHostContribution that discovers available agents from the
agent host process at startup. Each IAgent backend now exposes
an IAgentDescriptor with display metadata and auth requirements.
- Add IAgentDescriptor interface and listAgents() to IPC contract
- CopilotAgent/ClaudeAgent return descriptors via getDescriptor()
- Single AgentHostContribution discovers + registers dynamically
- Remove agentHostConstants.ts (no more hardcoded session types)
- AgentHostSessionListController/LMProvider take params instead
- Rename AgentSessionProviders.AgentHost -> AgentHostCopilot
- Update architecture.md, sessions.md, backlog.md
(Written by Copilot)
* Fix review findings: proxy, disposal, filtering, tests
- Add listAgents() forwarding to AgentHostServiceClient
- Guard async discovery against disposal race
- Add provider field to IAgentModelInfo for per-provider filtering
- Filter models and sessions by provider in LM provider and list
controller
- Update tests for new dynamic API and agent-host-copilot scheme
(Written by Copilot)
* Use DI for AgentHostLanguageModelProvider
(Written by Copilot)
* Strip @img/sharp native binaries from builds
sharp is a transitive dependency of the Claude Agent SDK used for
image processing. Its native .node binaries cause dpkg-shlibdeps
errors during Debian packaging due to $ORIGIN RPATH references.
Strip all @img/sharp-* platform packages since the agent host
doesn't need image processing at runtime.
(Written by Copilot)
* Strip Claude SDK vendored ripgrep binaries
The Claude Agent SDK bundles ripgrep binaries for all platforms
under vendor/ripgrep/. Wrong-architecture binaries cause macOS
Mach-O verification to fail. Strip them entirely via .moduleignore
(VS Code has its own ripgrep) and add to verify-macho skip list.
(Written by Copilot)
* Add tests for AgentSession, AgentService dispatcher, and workbench agent host components
(Written by Copilot)
* Add trace logging, IPC output channel, tool permissions, and attachment context
- Add Agent Host IPC output channel (only registered at trace log level) that
logs all IPC method calls, results, and progress events with full JSON payloads
- Add trace-level logging in AgentService dispatcher for all method calls
- Add trace-level logging in session handler for all progress events and session
resolution
- Wire up onPermissionRequest handler on CopilotClient.createSession and
resumeSession to auto-approve tool permission requests
- Add IAgentAttachment type to IPC contract and thread attachments from chat
variables (file, directory, selection) through sendMessage to the Copilot SDK
(Written by Copilot)
* Add tests for attachment context conversion and threading
(Written by Copilot)
* Add gap analysis docs for Copilot and Claude SDK implementations
(Written by Copilot)
* Sanitize env vars for Copilot CLI subprocess
Strip VSCODE_*, ELECTRON_* (except ELECTRON_RUN_AS_NODE), NODE_OPTIONS, and
other debug-related env vars that can interfere with the Node.js process the
SDK spawns. Matches the env sanitization from the extension implementation.
Also set useStdio and autoStart for proper CLI communication.
(Written by Copilot)
* Add error, usage, and title_changed event types to IPC contract
Add IAgentErrorEvent, IAgentUsageEvent, and IAgentTitleChangedEvent to the
progress event union. Wire up session.error and assistant.usage events from
the Copilot SDK to fire as IPC events instead of only logging. Handle error
events in the renderer session handler by rendering the error message. Usage
and title_changed events are logged at trace level.
(Written by Copilot)
* Add abortSession IPC method for proper cancellation
Add abortSession(session) to the IPC contract, implemented across AgentService,
CopilotAgent (calls session.abort()), ClaudeAgent (no-op, uses AbortController),
and the renderer proxy. Wire up cancellation in the session handler to call
abortSession before finishing, so the SDK actually stops processing.
(Written by Copilot)
* Address reviewer feedback: error finishes request, Claude abort, tests
- Error events now call finish() so the request doesn't hang if the SDK
doesn't send idle after an error
- ClaudeAgent.abortSession calls ClaudeSession.abort() which signals the
AbortController and creates a new one for future turns
- Add test: cancellation calls abortSession on the agent host service
- Add test: error event renders message and finishes the request
- Remove stale TODO in interruptActiveResponseCallback
- Use timeout() helper instead of raw setTimeout in test
- Update gap docs to reflect completed work
(Written by Copilot)
* Add permission request IPC round-trip (Written by Copilot)
* Remove Claude agent from agent-host process
Strip the Claude Agent SDK integration from the agent-host utility process
to focus on the Copilot SDK path.
- Delete src/vs/platform/agent/node/claude/ (claudeAgent, claudeSession, claudeToolDisplay)
- Remove @anthropic-ai/claude-agent-sdk from package.json
- Remove AgentHostClaude enum member and all switch cases
- Remove Claude command registration in electron-browser chat.contribution
- Clean up build scripts (.moduleignore, verify-macho, gulpfile.vscode)
- Narrow AgentProvider type to just 'copilot'
- Update tests and documentation
(Written by Copilot)
* Wire up permission confirmation UI with ChatToolInvocation (Written by Copilot)
* Fix reviewer feedback: safe permission serialization, deny on abort/dispose (Written by Copilot)
* Forward reasoning events as thinking blocks (Written by Copilot)
* Pass workspace folder as workingDirectory to Copilot SDK (Written by Copilot)
* Store and pass workingDirectory on session resume, update gap docs (Written by Copilot)
* Fix permission rendering, session-scoped permissions, and test gaps (Written by Copilot)
* Auto-approve read permissions inside workspace folder (Written by Copilot)
* Move read auto-approve into CopilotAgent where permission policy belongs (Written by Copilot)
* Update gap docs (Written by Copilot)
* Use log language for IPC output channel, add trace prefix (Written by Copilot)
* Add tool rendering gaps to docs (Written by Copilot)
* Stringify URIs in IPC output channel for readability (Written by Copilot)
* Fix IPC output channel: use log languageId with non-log channel for proper append + syntax highlighting (Written by Copilot)
* Fix build errors: add URI import, fix test mock types (Written by Copilot)
* Don't localize agent host provider strings (Written by Copilot)
* Remove claude-agent-sdk from eslint allowed imports (Written by Copilot)
* fix test
* initial thoughts
* Rename folder to agentHost
* Fix paths
* Fixes
* Fixes for copilot
* Fix moduleignore
* first working protocol version
align more closely with protocol
json rpc and some gaps
* cleanup
* Fix copilot pty.node packaging
* Fix test
* prebuild packaging
* Agenthost server fixes
* Update monaco.d.ts
* Update docs
* Fixes
* Build fix
* Fix build issues
* reduce duplication in side effecting code
* fix model switching not working
* reduce mock duplication
* Build fixes
* Copy vscode's node.pty
* And ripgrep
* And thsi
* Ripgrep goes to non-SDK
* Skip copy for stable build
* Remove outdated script
* Build fixes for asar
* fix
* Add some logging
* Fix for windows
* Fix
* Logs
* build: add glob diagnostic for copyCopilotNativeDeps
* build: check both node_modules/ and .asar.unpacked/ for source binaries
* Fix
* Remove excalidraw
---------
Co-authored-by: Connor Peet <connor@peet.io>
Co-authored-by: Connor Peet <copeet@microsoft.com>
* Fix PSReadline for screen reader, disable header
* Add psreadline to index.ts for build
* organize much better
* compile
* list out exact folder
* try if codesign is the problem
* Update comment to be more accurate
* Undo changes for noLogo
The new esbuild-based CI pipeline (core-ci) uses curated resource
patterns in build/next/index.ts to copy non-JS assets into the bundle
output. When built-in .prompt.md files were added for the sessions
window, they were included in the legacy pipeline's
vscodeResourceIncludes (build/gulpfile.vscode.ts) but not in the
desktopResourcePatterns used by the esbuild pipeline.
This caused the prompt files to be missing from release builds
(out-vscode-min), even though they worked correctly when running
from sources (where copyAllNonTsFiles copies everything).
Add 'vs/sessions/prompts/*.prompt.md' to desktopResourcePatterns
to match the existing entry in vscodeResourceIncludes.
- Introduced adjustments for source maps in the NLS plugin to ensure accurate mapping after placeholder replacements.
- Implemented deferred processing for source maps to handle edits more effectively, preserving unmapped segments.
- Updated tests to validate column mappings and ensure correctness in both minified and non-minified builds.
- Improved documentation to reflect changes in source map generation and adjustments.
* chore: ralph loop checkpoint - 2026-02-17 12:48
* fix shell script
* Inline compile stage into platform jobs
Remove the standalone Compile stage from the ADO pipeline. Each
platform job (Windows, Linux, macOS, Alpine, Web) now compiles
TypeScript itself instead of downloading a shared Compilation artifact.
- Add VSCODE_RUN_CHECKS parameter to Linux jobs for hygiene/lint/CG
- Add VSCODE_RUN_COMPILE_EXTRAS parameter to macOS jobs for telemetry
extraction and sourcemap upload
- Remove VSCODE_COMPILE_ONLY parameter entirely
- Delete product-compile.yml (no longer referenced)
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Always run telemetry extraction on macOS builds
The macOS Universal app merge requires both x64 and arm64 builds to have
identical file sets. Telemetry extraction was only running on arm64 (via
VSCODE_RUN_COMPILE_EXTRAS), causing the universal merge to fail due to
missing telemetry-core.json and telemetry-extensions.json in the x64 build.
Move telemetry extraction outside the VSCODE_RUN_COMPILE_EXTRAS gate so
it runs on all macOS builds. Sourcemap upload remains gated.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Run telemetry extraction on all client builds
All client builds (Linux, Windows, macOS) need telemetry-core.json and
telemetry-extensions.json. Previously only macOS arm64 ran extract-telemetry.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Only run telemetry extraction on Linux x64 (client build)
Linux arm64 and armhf are server-only builds, no need for telemetry files.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Remove .ralph scaffolding files
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Fix telemetry extraction on Windows with native PowerShell
The bash extract-telemetry.sh script fails on Windows because the Unix
bin shim gets interpreted as Node.js code. Use a native PowerShell
implementation that calls the extractor JS entry point directly.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Handle missing telemetry files gracefully on Windows
The telemetry extractor may skip emitting declarations-resolved.json
when no events are found. Handle this case with a warning instead of
failing the build.
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Use deterministic build date from git commit timestamp
When each platform compiles independently, the build date embedded in
cli.js (via INSERT_PRODUCT_CONFIGURATION) differs between machines
because each uses new Date().toISOString(). This causes the macOS
Universal app merge to fail since cli.js SHA differs between x64/arm64.
Fix: use the git commit date (git log -1 --format=%cI HEAD) instead
of the current wall-clock time. This ensures all independent builds
on different machines produce identical timestamps.
Updated in:
- build/lib/date.ts: writeISODate() uses git commit date
- build/next/index.ts: bundle(), transpile, and readISODate fallback
all use git commit date
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* Remove redundant extensions-ci task from CI pipelines
core-ci (esbuild path) already includes cleanExtensionsBuildTask,
compileNonNativeExtensionsBuildTask, and compileExtensionMediaBuildTask.
Running extensions-ci in parallel caused a race condition where
core-ci's rimraf of .build/extensions clashed with extensions-ci
writing to the same directory.
Also removes dead code:
- extensions-ci and extensions-ci-pr task definitions (fully subsumed)
- core-ci-old task (useEsbuildTranspile is always true)
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
* remove VSCODE_RUN_COMPILE_EXTRAS
* address PR feedback regarding code duplication of `getGitCommitDate()` function by exporting a single helper function from `build/lib/date.ts` and importing it in `build/next/index.ts` to ensure consistent behavior and improve code maintainability.
* update readISODate function to return git commit date instead of current date
* add telemetry sorting script and integrate into build process for consistent output
* refactor telemetry extraction process: replace shell script with TypeScript implementation
* update skill
* update telemetry-extractor dependency to version 1.19.0
* fix build
* fix more duplicate telemetry definition issues
* cleanup
* refactor: consolidate validation checks into quality checks and remove obsolete tasks
* bust the cache
* undo cache bust
* fix expression
* fix
* fix: update Azure storage account name in quality checks
* fix: initialize stages set with 'Quality'
* fix: add VSCODE_BUILD_TYPE parameter with options for Product and CI builds
* fix: update Azure Pipeline CLI to use parameters instead of variables for queueing builds
* fix: update VSCODE_BUILD_TYPE parameter values for clarity
* fix: update default value for VSCODE_BUILD_TYPE parameter to 'Product'
* leaner
* even leaner
* only npm ci in build
* 💄
* run entire npm ci
* fix
* fix
* fix it
* Inline CompileCLI into platform stages
- Remove centralized CompileCLI stage
- Move CLI jobs into Windows, Linux, macOS stages as independent jobs
- CLI jobs now compile, publish unsigned mid-job, sign, and publish signed
- Platform compile jobs use deemon + waitForArtifacts for async CLI download
- Delete separate CLI sign jobs (now merged into CLI compile jobs)
- Remove CompileCLI from publish.ts stage tracking
* fix: macOS CLI signing - use proper directory structure for ESRP
* fix: add BUILDS_API_URL to Windows and Linux job templates
* fix: label Linux CLI jobs clearly
---------
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
Co-authored-by: João Moreno <22350+joaomoreno@users.noreply.github.com>
* feat: enhance source map handling in NLS and private field conversion
- Implemented inline source maps in the NLS plugin to ensure accurate mapping from transformed source back to original.
- Modified `transformToPlaceholders` to return edits for source map adjustments.
- Added `adjustSourceMap` function to update source maps based on text edits in `convertPrivateFields`.
- Introduced tests for source map accuracy in both NLS and private field conversion scenarios.
- Updated documentation to reflect changes in source map handling and the rationale behind accepting column imprecision.
* address PR review: fix source name collision, guard unmapped segments, reorder mangle before NLS
- Implemented copyAllNonTsFiles() for transpile/dev builds to copy all non-TS files from src/ to output, aligning with previous gulp behavior.
- Retained copyResources() for production builds using curated resource patterns.
- Removed unnecessary devOnlyResourcePatterns and testFixturePatterns.
- Updated watch mode to accept any non-TS file changes.
