@bastani/atomic 0.9.19-alpha.1 → 0.9.19-alpha.2

package/dist/builtin/subagents/skills/herdr/SKILL.md

@@ -0,0 +1,202 @@
+---
+name: herdr
+description: "Control Herdr, a terminal multiplexer for coding agents. Use only when the user explicitly mentions Herdr or asks to use Herdr to inspect or control panes, tabs, workspaces, commands, or another agent. Do not use merely because a task could benefit from a background terminal, delegation, or parallel work. Requires HERDR_ENV=1."
+---
+
+# Herdr
+
+Herdr organizes terminals into workspaces, tabs, and panes, recognizes coding agents running inside panes, and exposes the current session through the `herdr` CLI.
+
+Before issuing any control command, verify that this agent is running inside a Herdr-managed pane:
+
+```bash
+test "${HERDR_ENV:-}" = 1
+```
+
+If the check fails, say that you are not running inside Herdr and stop. Do not inspect or control the focused Herdr session from outside Herdr.
+
+When the check passes, the `herdr` binary in `PATH` talks to the current session. Use it to inspect neighboring work, create terminal layout, start agents and commands, read output, and wait for state changes.
+
+## Learn the current CLI
+
+The installed binary is the authority for command syntax. Start with:
+
+```bash
+herdr --help
+```
+
+Then print the relevant command group by running the group without a subcommand:
+
+```bash
+herdr agent
+herdr pane
+herdr workspace
+herdr tab
+herdr worktree
+herdr terminal
+herdr notification
+herdr integration
+herdr session
+herdr machine
+```
+
+Do not run bare `herdr` for discovery; it launches or attaches the TUI. Do not probe a mutating nested command by omitting arguments. Commands such as `herdr workspace create` are valid with defaults and will execute.
+
+Most control commands return JSON. Read identifiers and state from those responses instead of predicting them.
+
+## Understand layout, panes, and agents
+
+Choose the primitive that matches the job:
+
+- Workspace, tab, and pane topology organize terminal locations.
+- Pane commands control raw terminals, shells, tests, servers, input, and output.
+- Agent commands control the recognized coding agent currently occupying a pane.
+
+A pane exists whether or not it contains an agent. `agent start` requires an existing available shell pane and never creates, splits, or moves layout. Use pane commands for ordinary processes. Use agent commands when Herdr must validate agent identity or interpret `idle`, `working`, `blocked`, `done`, and `unknown` lifecycle states.
+
+Agent commands accept either a unique live agent name or the pane ID currently hosting that agent. They do not accept terminal IDs or bare agent-kind labels. Names must match `[a-z][a-z0-9_-]{0,31}` and be unique among live agents. A name follows the current pane occupant and is cleared when that agent exits, is released, or is replaced.
+
+`idle` and `done` both mean the agent is ready for input. The CLI/API uses the server's seen state to distinguish them; explicit focus commands mark the target seen, while reads do not. Each TUI client tracks viewed completions independently, so its Done badge can differ from the CLI or another client's badge. `blocked` means Herdr recognized an approval or question UI. `unknown` means an agent is present but Herdr cannot classify it confidently; it does not prove completion.
+
+## Use IDs and caller context
+
+Public IDs are opaque stable handles:
+
+- workspace: `w1`
+- tab: `w1:t1`
+- pane: `w1:p1`
+
+Closed tab and pane IDs are not reused. A pane moved into another workspace receives a new workspace-qualified pane ID. After `pane move`, continue with `.result.move_result.pane.pane_id` or the live agent name. The old value is reported as `.result.move_result.previous_pane_id`; only the moved process's inherited caller context keeps resolving that old ID, so do not use it as a general agent target.
+
+Herdr injects the caller's context into each managed pane:
+
+```bash
+printf '%s\n' "$HERDR_WORKSPACE_ID" "$HERDR_TAB_ID" "$HERDR_PANE_ID"
+```
+
+Prefer `--current` when a pane command should target the calling pane. Omitting a target may use the UI-focused pane, which can belong to the user or another client.
+
+Discover live state with:
+
+```bash
+herdr workspace list
+herdr tab list --workspace "$HERDR_WORKSPACE_ID"
+herdr pane current --current
+herdr pane list --workspace "$HERDR_WORKSPACE_ID"
+herdr agent list
+```
+
+Creation responses expose the IDs to use next. `workspace create` returns `.result.workspace`, `.result.tab`, and `.result.root_pane`. `tab create` returns `.result.tab` and `.result.root_pane`. `pane split` returns the new pane as `.result.pane`.
+
+IDs and live agent names are scoped to one server. Two saved SSH machines can both have `w1:p1` or an agent named `reviewer`. Selecting a machine in the TUI does not retarget commands running in your pane: they still use the inherited session and socket context. Run remote control commands on the intended host with its explicit session, and rediscover IDs there.
+
+`herdr machine list` lists saved connection profiles, not a cross-machine pane inventory; add `--json` for scripts. Only add, remove, enable, or disable profiles when the user asks. Removing a profile disconnects the client but does not stop remote sessions. Adding a machine uses the remote default session unless `--remote-session` is explicitly supplied. Setup asks before stopping an incompatible server and defaults to No; do not approve replacement without the user's consent. Experimental handoff is not part of `machine add`.
+
+## Start and coordinate an agent
+
+Default to a sibling pane in the current tab and the current working directory. Do not create a workspace, tab, worktree, or different cwd unless the user explicitly requests that topology or location.
+
+Honor a direction requested by the user. Otherwise inspect the caller pane:
+
+```bash
+herdr pane layout --pane "$HERDR_PANE_ID"
+```
+
+Split a wide pane to the right and a narrow or tall pane down. Avoid repeated same-direction splits that create unusably narrow columns or short rows. Keep the user's focus in the calling pane and explicitly preserve the caller's working directory:
+
+```bash
+herdr pane split --current --direction right --cwd "$PWD" --no-focus
+```
+
+Replace `right` with `down` when appropriate. Read the new pane ID from `.result.pane.pane_id`.
+
+An available shell pane must be at its interactive prompt, with the shell itself in the foreground and no foreground command, editor, or agent running. Start a supported agent in that pane with a useful unique name:
+
+```bash
+herdr agent start reviewer --kind codex --pane <returned-pane-id>
+```
+
+Use the kind requested by the user. Run `herdr agent` to inspect the installed kind list and options. Pass native agent arguments only after `--`:
+
+```bash
+herdr agent start reviewer --kind codex --pane <returned-pane-id> -- <agent-args...>
+```
+
+A successful `agent start` returns only after Herdr detects the expected agent in the same pane and considers it ready for interactive input. If the agent is blocked during startup, the command returns `agent_not_ready` immediately but keeps the name available for `agent read` and `agent send-keys`. Wait until the agent becomes idle before prompting it. Startup defaults to a 30-second timeout.
+
+Submit work through the agent surface:
+
+```bash
+herdr agent prompt reviewer "Review the current diff and report only actionable findings." --wait --timeout 120000
+```
+
+`agent prompt` honors the pane's live bracketed-paste mode and sends text followed by encoded Enter as one ordered submission. It reports successful submission only after both have been written; that alone does not prove the agent started a turn. The submit delay grows with prompt size for Codex on Windows. It rejects an agent already waiting at an approval or question dialog with `agent_blocked` before sending any input. Inspect the blocked UI and ask the user before answering it. For normal agent work, `--wait` is enough: it waits for the first settled `idle`, `done`, or `blocked` state. Do not repeat those defaults with `--until`.
+
+With `--wait`, a prompt sent from a non-working state must produce observed `working` or `blocked` activity. After submission, Herdr waits up to five seconds for that activity; unrelated `idle`, `done`, or session changes do not satisfy this gate. It returns `agent_prompt_stalled` if no activity is observed, or `timeout` if the caller's timeout expires first. The caller timeout includes submission time. Without a timeout, the settled-state wait is indefinite after activity is observed. This wait tracks lifecycle state, not an individual turn; if the agent is already working, completion of the active turn may satisfy it.
+
+Use `--until` only for a state-specific workflow, such as waiting for an already-running agent to request input:
+
+```bash
+herdr agent wait reviewer --until blocked --timeout 120000
+```
+
+Without `--until`, standalone `agent wait` uses the same settled-state defaults as `agent prompt --wait`.
+
+Use logical keys for interactive agent UI controls:
+
+```bash
+herdr agent send-keys reviewer esc
+herdr agent send-keys reviewer ctrl+c
+```
+
+Herdr validates all keys before writing any bytes. Read the result through the resolved agent:
+
+```bash
+herdr agent get reviewer
+herdr agent read reviewer --source recent-unwrapped --lines 120
+```
+
+If a wait fails or returns `blocked`, inspect `agent get` and `agent read` before deciding what input to send. A timeout or stalled response does not prove the prompt was never delivered; do not blindly submit it again. Use the pane surface only when raw terminal control is intentional.
+
+## Run an ordinary command in another pane
+
+Create a sibling pane with the same geometry rule, preserve the caller's working directory, and keep user focus unchanged:
+
+```bash
+herdr pane split --current --direction right --cwd "$PWD" --no-focus
+```
+
+Read the new pane ID from `.result.pane.pane_id`, then run and inspect the command:
+
+```bash
+herdr pane run <returned-pane-id> "just test"
+herdr pane wait-output <returned-pane-id> --match "test result" --timeout 120000
+herdr pane read <returned-pane-id> --source recent-unwrapped --lines 120
+```
+
+`pane run` atomically sends command text and Enter. `pane wait-output` searches the selected snapshot immediately, so output that already exists can match. Use `--match <text>` for a literal substring or `--regex <pattern>` for a Rust regular expression. Omitting `--timeout` allows an indefinite wait.
+
+Use the read source that matches the task:
+
+- `visible`: the currently rendered viewport.
+- `recent`: recent rendered output, including soft wraps.
+- `recent-unwrapped`: recent output with soft wraps joined; prefer it for logs and transcripts.
+- `detection`: the plain-text bottom-buffer snapshot used for agent detection.
+
+Use `--format ansi` when colors and terminal styling are evidence. Otherwise use text.
+
+`--lines` asks Herdr for more rows from the pane's available screen and host scrollback. If increasing it does not reveal more of a completed response, the pane is probably running the agent on the terminal's alternate screen. Rows that leave the alternate screen do not enter Herdr's host scrollback, so a larger line count cannot recover them.
+
+After that failed read, ask the agent to write its complete response as Markdown in a temporary directory and reply only with the file path, then read the file directly. Use this only as a fallback; do not request file output in the initial prompt.
+
+## Safety and coordination rules
+
+- Use `--no-focus` for background work unless the user asked to switch context.
+- Use `--current`, an explicit pane ID, or a unique agent name. Do not rely on another client's focused pane.
+- Parse IDs from JSON responses. Do not derive them from sidebar order or examples.
+- Do not close workspaces, tabs, panes, or sessions you did not create unless the user explicitly asked. `workspace close --group` closes the primary workspace and its linked worktree workspaces; never add it merely to bypass `workspace_group_close_required`.
+- Use `--trust-repository` only after the user has verified the repository. It grants per-request Git trust; it is not a routine retry for a failed worktree command.
+- Client and server versions can differ after an update. Check `herdr status` before relying on new server features. A missing method is not permission to stop or upgrade a server.
+- Never run `herdr server stop` from an active session unless the user explicitly intends to stop the server and its pane processes.
+- Never kill the main Herdr process. Use named test sessions for experiments that need an isolated server.
+- CLI server errors are JSON on stderr with exit status 1. CLI syntax errors exit with status 2.

package/dist/builtin/subagents/skills/subagent/SKILL.md

@@ -207,9 +207,13 @@ Avoid duplicate output paths in parallel tasks. Concurrent children should not w
 Concurrent writers conflict. `code-simplifier` and `debugger` change files. Do not run two writers in parallel against the same worktree unless you isolate them with `worktree: true`.
 
 
-### Foreground execution and fresh follow-ups
+### Foreground, background, and automatic yielding
 
-All subagent execution runs in the foreground and returns its result to the parent call. Parallel tasks may still run concurrently within one foreground invocation, and forked context still creates branched child sessions.
+Choose the observation mode for each authorized task. No extra user confirmation is needed merely to choose foreground or background. In owner-bound main and workflow-stage sessions, omitted `wait` or `wait: { kind: "background" }` returns after admission. Use `wait: { kind: "foreground", budgetMs: 30000 }` when the result is needed next. If the observation budget expires, the same child keeps running in the background; do not relaunch it. This applies to single and parallel calls.
+
+Use `subagent({ action: "wait", id: taskId, budgetMs: 1000 })` to observe an existing task, `status` to inspect its state, or `interrupt` to stop it. A yielded receipt is not a terminal result. Background counts stay below the prompt; `/tasks` opens inspection only on command. A shaded completion notification reaches the owning chat without requiring a model reply. A later wait does not extend the owner's lifetime.
+
+Intercom `ask` cannot revive a completed, failed, interrupted, or cancelled noninteractive child, even if its retained registration says `idle`. New asks fail immediately, and termination fails an already-admitted ask that has not received a reply. Use a fresh child for follow-up work. Live interactive idle sessions and workflow post-mortem conversations remain separate reply-capable cases; `send` transport behavior is unchanged.
 
 Completed, interrupted, and parent-question children are terminal for continuation. Do not address a prior child or sibling set by run ID. Start follow-up work with the normal launch form and an explicit handoff:
 
@@ -408,9 +412,9 @@ If a prompt-template extension is installed, additional user prompt templates ca
 
 ## Best Practices
 
-### Choose foreground intentionally
+### Choose observation intentionally
 
-Use foreground runs for every delegated call so the result gates the parent's next action. Do not duplicate a delegated job while waiting.
+Use background observation for independent work and foreground-first observation when the next action depends on the result. If foreground observation yields, wait for the existing task's actual completion before consuming its result. Do not duplicate delegated work or mistake a launch receipt for a finished report. The owner's usual agent observation budget is 30 seconds, independent of execution lifetime.
 
 ### Keep writes single-threaded by default
 
@@ -451,12 +455,12 @@ Use `/name` so intercom targeting stays stable.
 
 ### Locate, analyze, fix
 
-Use explicit follow-up calls when each result should guide the next task:
+When each result guides the next task, wait for its terminal completion before starting the dependent step:
 
 ```typescript
-const context = await subagent({ agent: "codebase-locator", task: "Map the auth files and tests relevant to: ..." });
-const analysis = await subagent({ agent: "codebase-analyzer", task: "Trace current behavior of the mapped files. Use the returned context: ..." });
-await subagent({ agent: "debugger", task: "Reproduce the failure and patch the root cause. Use the returned analysis: ..." });
+subagent({ agent: "codebase-locator", task: "Map the relevant auth files.", wait: { kind: "foreground" } })
+// If yielded, observe the returned task ID or await its completion notice.
+// Only then give the result to codebase-analyzer, and later to debugger.
 ```
 
 ### Clarify → Discover → Implement → Review (self-orchestrated workflow)
@@ -474,7 +478,7 @@ clarify when needed → validation contract → optional bounded discovery → o
 
 The validation contract defines completion before code is written: expected behavior, checks, commands or user flows to exercise, and evidence the writer should return. Keep it lightweight for small tasks, but make it explicit enough that reviewers and validators are checking the intended outcome rather than the writer’s own assumptions. Subagent runs do not carry a structured `acceptance` field, infer acceptance policies, inject acceptance-report prompts, or run acceptance gates; put any evidence requirements directly in the task text. Do not set removed acceptance config fields on `subagent()` calls, parallel task items, or agent frontmatter; move those requirements into the assigned task text instead.
 
-The first writer implements the approved change. The parent waits for its foreground handoff before review, and does not make parallel edits to the same worktree. Treat the writer handoff as the transition into review, not as final completion, unless the user explicitly asked for writer-only work, review-only output, or to stop after implementation. Specialist reviewers inspect the resulting diff from fresh context when warranted. The final fix writer applies synthesized fixes, then the parent looks over the final diff before completing. Ask only needed questions before a non-interactive launch.
+The first writer implements the approved change. The parent waits for its terminal handoff before review, even if the foreground observation has yielded, and does not make parallel edits to the same files. Treat the handoff as the transition into review unless the user requested writer-only work. Reviewers inspect the resulting diff from fresh context. The fix writer applies accepted findings, then the parent checks the final diff. Ask only needed scope questions before a non-interactive launch.
 
 For complex or risky changes, increase review and validation fanout when user intent or correctness risk materially warrants it rather than automatically trusting one reviewer. Use distinct angles such as correctness/regressions (`codebase-analyzer`), failure-mode hunt (`debugger` inspect-only), pattern fit (`codebase-pattern-finder`), prior-decision conformance (`codebase-research-*`), and external-spec conformance (`codebase-online-researcher`). When reviewers find non-trivial issues or the fix writer touches many lines, consider another focused review round before final validation.
 
@@ -528,7 +532,7 @@ subagent({
 
 When implementation review is part of the requested shape, do not treat the first review as the final step: synthesize findings against user scope and the validation contract, then launch one writer for accepted fixes when implementation is authorized.
 
-When a writer completes, treat its handoff as an intermediate state when review is part of the requested shape. The next parent action is bounded review, then synthesis, then a fix writer if reviewers found fixes worth doing now. Keep these calls in the foreground so each handoff is available before the next action.
+When a writer completes, treat its handoff as an intermediate state when review is part of the requested shape. The next action is review, then synthesis and a fix writer if needed. Foreground-first observation is convenient, but a yield requires waiting for the original task's terminal handoff before the next dependent action.
 
 When the user explicitly asks to keep reviewing until the work is clean, repeat writer → fresh-specialist-reviewers → synthesized-fix-writer cycles until reviewers find no blockers or fixes worth doing now, remaining feedback is optional or intentionally deferred, an unapproved product/scope/architecture decision needs the user, or the max review-round cap is reached. Default to 3 review rounds unless the user sets a different cap.
 

package/dist/builtin/subagents/skills/tmux/SKILL.md

@@ -52,7 +52,7 @@ metadata:
 
 # tmux
 
-Use for existing interactive tmux sessions. For one-shot commands, use normal shell. When a command must outlive a tool call, start it inside an interactive tmux session; Atomic tools do not provide background execution.
+Use tmux for interactive terminal sessions and terminal UI verification. For ordinary commands, use `bash`: supported owner-bound calls can run in the background explicitly with `wait: { kind: "background" }` or automatically after their foreground observation budget expires. Use tmux when you need a separate interactive terminal, not merely because a command outlives a tool call. Background tasks remain bound to their owning Atomic session.
 
 On Windows, use [psmux](https://github.com/psmux/psmux), the native Windows tmux-compatible multiplexer. psmux automatically provides the `tmux` alias, so keep using the `tmux` commands below; no command changes are needed.