The `/` completion of the New Task page only listed the commands the agent
offered in its last session, cached once for all projects, and matched them
by substring. Before a project's first session, or after a session in another
project, its skills were missing, and `/trig` never matched `/triage`.
- Remember the commands an agent offered per project root (settings stay shared).
- New `agent_commands` command: the agent's last list for the project (or the
task's own list in a session) merged with, for Claude Code, the skills
(`.claude/skills/*/SKILL.md`) and custom commands (`.claude/commands/**/*.md`,
nested as `dir:name`) found from the working directory upward and in
`~/.claude`. The agent's entries win on a name; skills with
`user-invocable: false` are left out. Discovery runs off the command thread.
- Sessions discover from the task's own directory (worktree or checkout).
- Command matching ranks prefix, plugin-name prefix, substring, then letters in order.
Claude-Session: https://claude.ai/code/session_01VUYsdXA2nkF5v6zsC93dpn
About 270 UI strings were JS template literals with interpolation passed to exact catalog lookup, so they could never be translated; about 25 more spliced lowercase English fragments into sentences.
`tx` now also works as a template tag (`` tx`Remove data source '${name}'?` ``): it looks up the entry by the literal parts (`{name}` placeholders), translates interpolated values that have entries, and remembers generated text so it re-translates after a language switch or when the catalog loads later. Call sites use the tag; spliced fragments are written as whole sentences per variant. Adds 280 Chinese translations and removes stale ones.
The catalog guard now also fails on untagged template literals reaching exact lookup (unless allowlisted with a reason), tagged text missing from the catalog, case-changed or ternary-suffix interpolations, and duplicate template patterns.
Claude-Session: https://claude.ai/code/session_01VUYsdXA2nkF5v6zsC93dpn
## What
An Agents feature: hand a task to a coding agent and follow it from the IDE. Agents speak the [Agent Client Protocol](https://agentclientprotocol.com). Built in: Claude Code, Codex, Gemini CLI, GitHub Copilot, Cursor, OpenCode, Qwen Code, Kimi CLI, Goose, Auggie, Kiro, Mistral Vibe. You can add your own in `<config>/agents/<id>.toml` (`name`, `command`, `args`, `[env]`). An agent whose command isn't on `PATH` is listed under "Not installed".
- **New Task page.** An editor page with a prompt box. You pick the agent, model, mode and thinking effort, and whether it runs in a new worktree (from a chosen branch) or in the checkout.
- **Agents tool window.** Tasks are grouped by what they need: Needs Input, Failed, Running, Ready to Review, Archived. It also shows a stripe badge, a status bar item and notifications.
- **Session tab.**
- The conversation: Markdown, thinking, tool calls with diff stats, the plan.
- Permission requests appear as an editor banner.
- The changed files are compared against the task's base commit and open in the diff viewer.
- A context-usage meter.
- **Composer.**
- `@` references files from the project index. A referenced file's text is sent when the agent accepts embedded context, otherwise a `resource_link`.
- `/` lists the agent's commands; ↑/↓ recall earlier messages.
- Images can be pasted, dropped or picked; Shift+Tab cycles the mode.
- While the agent works, Enter queues the message and ⌘↵ interrupts the turn and sends it now. Queued messages can be sent now, edited or removed.
- **Send to Agent.** In the editor and Project view context menus. It adds `@file:10-20` references to the message you typed last.
- Landing the work is left to the existing Git features: the worktree is a normal branch.
## How
- **`crates/ide-agent`**: one ACP agent process per task, running in the task's directory.
- It folds `session/update` into a timeline and answers `session/request_permission`.
- It tracks the session's settings: `configOptions`, falling back to the older `modes` / `models` fields. The user's choices are applied again on every new or reloaded session.
- Tasks are saved as JSON per task. Sessions resume with `session/load` without duplicating history, and old task files are upgraded when read.
- **`ide-jsonrpc`**: adds newline-delimited framing (ACP) next to Content-Length (LSP).
- **`ide-vcs`**: adds `numstat_with_worktree` for the changes list.
- **Editor**:
- The Switcher (⌃⇥) lists every open editor, as IDEA does, not just files.
- `EditorKindContribution.restorable` lets a kind's tabs be saved with the editor session, so agent tabs survive reloads.
- **UI**: `Select` popups now size to fit short option details (they used to cut them off).
## Testing
- `cargo test -p ide-agent` (14 tests). They use an in-memory ACP agent and cover permissions, restart with `session/load`, settings, the queue with send-now and stop, and prompt blocks.
- There is also an ignored end-to-end test against the real Claude Code adapter: `OPENIDEA_AGENT_E2E=1 cargo test -p ide-agent real_claude -- --ignored`.
- `pnpm vitest run`: 2052 tests.
- `cargo clippy -p ide-agent -p openidea --all-targets -- -D warnings` and typecheck are clean.
- Driven in the running app with Claude Code:
- A task in a worktree, with permission answers, changes listed and the diff opened.
- A follow-up after restarting the IDE.
- `@` reference with embedded content, `/` commands, Shift+Tab mode changes.
- A queued message and send-now, an image prompt, Send to Agent from a selection.
- The Switcher and restored tabs after a reload.
Not covered yet: only Claude Code was driven end to end. The other agents go through the same ACP path but weren't run here.
Merged on the maintainer's go-ahead. Merge with main: kept both sides in Cargo.toml (ide-agent + ide-clipboard), the feature registry (agents + vim) and zh.json (no duplicate keys). After the merge: typecheck clean, full vitest 268 files / 2247 tests passed, ide-agent/ide-jsonrpc/ide-vcs 144 passed, bindings unchanged after regeneration.
l0ng-ai