1.0.22

2026-04-09

• MCP tools with non-standard JSON schemas are now sanitized for compatibility with all model providers
• Better handling of large images from MCP and extension tools
• Improved rendering performance with a new simplified inline renderer
• Show a clear message to contact your organization administrator when remote sessions are blocked by policy
• Sub-agent activity no longer shows duplicated tool names (e.g. "view view the file...")
• Permission checks and other hooks now work correctly when using Anthropic models via BYOM/BYOK configuration
• Slash command picker appears above the text input for a more stable layout
• Custom agents can now declare a skills field to eagerly load skill content into agent context at startup
• Plugins can now display a post-install message with setup instructions after installation
• Remove .vscode/mcp.json and .devcontainer/devcontainer.json as MCP server config sources; CLI now only reads .mcp.json. A migration hint appears when .vscode/mcp.json is detected without .mcp.json.
• Plugins remain enabled across sessions and auto-install on startup based on user config
• Add sub-agent depth and concurrency limits to prevent runaway agent spawning
• Warn when resuming a session that is already in use by another CLI or application
• CLI no longer crashes on systems affected by a V8 engine bug in grapheme segmentation
• sessionStart and sessionEnd hooks fire once per session in interactive mode instead of once per prompt
• Plugin agents respect the model specified in their frontmatter

1.0.22-0

Added

• Add sub-agent depth and concurrency limits to prevent runaway agent spawning
• Warn when resuming a session that is already in use by another CLI or application

Fixed

• CLI no longer crashes on systems affected by a V8 engine bug in grapheme segmentation
• sessionStart and sessionEnd hooks fire once per session in interactive mode instead of once per prompt
• Plugin agents respect the model specified in their frontmatter

1.0.21

2026-04-07

• Add copilot mcp command for managing MCP servers
• Spinner no longer appears stuck when a long-running async shell command is active
• Enterprise GitHub URL input in the login flow now accepts keyboard input and submits on Enter
• Slash command picker no longer flickers or shifts the input while filtering
• Timeline no longer goes blank when content shrinks (e.g., after cancelling or tool completion)
• Plan mode timeline display shows user text without a redundant "Plan" prefix
• Reduce memory usage by automatically shutting down shell sessions that are no longer needed
• Hooks configured with PascalCase event names now receive VS Code-compatible snake_case payloads with hook_event_name, session_id, and ISO 8601 timestamps

1.0.20

2026-04-07

• Add copilot help monitoring topic with OpenTelemetry configuration details and examples
• Spinner stays active until background agents and shell commands finish, and user input remains available throughout
• Azure OpenAI BYOK defaults to the GA versionless v1 route when no API version is configured
• Reduce UI sluggishness during live response streaming
• /yolo and --yolo now behave identically and /yolo state persists across /restart

1.0.20-1

Pre-release 1.0.20-1

1.0.20-0

Added

• Add copilot help monitoring topic with OpenTelemetry configuration details and examples

Improved

• Spinner stays active until background agents and shell commands finish, and user input remains available throughout
• Azure OpenAI BYOK defaults to the GA versionless v1 route when no API version is configured
• Reduce UI sluggishness during live response streaming
• /yolo and --yolo now behave identically and /yolo state persists across /restart

1.0.19

2026-04-06

• /mcp enable and /mcp disable now persist across sessions
• OpenTelemetry monitoring: subagent spans now use INTERNAL span kind, and chat spans include a github.copilot.time_to_first_chunk attribute (streaming only)
• Plugin hook scripts with missing execute permissions now run correctly on macOS
• Custom agent is properly restored when resuming a session where the agent display name differs from its filename
• Skip IDE auto-connect when session is already in use by another client
• Slash command timeline entries now include the command name (e.g., "Review", "Plan") for better context

1.0.19-0

Improved

• Skip IDE auto-connect when session is already in use by another client
• Slash command timeline entries now include the command name (e.g., "Review", "Plan") for better context

Fixed

• Plugin hook scripts with missing execute permissions now run correctly on macOS
• Custom agent is properly restored when resuming a session where the agent display name differs from its filename

1.0.18

2026-04-04

• New Critic agent automatically reviews plans and complex implementations using a complementary model to catch errors early (available in experimental mode for Claude models)
• Session resume picker correctly groups sessions by branch and repository on first use
• preToolUse hook permissionDecision 'allow' now suppresses the tool approval prompt
• Add notification hook event that fires asynchronously on shell completion, permission prompts, elicitation dialogs, and agent completion

1.0.17

2026-04-03

• Built-in skills are now included with the CLI, starting with a guide for customizing Copilot cloud agent's environment
• MCP OAuth flows now support HTTPS redirect URIs via a self-signed certificate fallback, improving compatibility with OAuth providers that require HTTPS (e.g., Slack)
• /resume session picker loads significantly faster, especially with large session histories