didericis/bot-bottle
1
0
Fork 0
Code Issues 7 Pull Requests 15 Actions Packages Projects Releases Wiki Activity

docs(prd-0021): rewrite as standalone — no references to closed PR #48
test / unit (pull_request) Successful in 18s
Details
test / integration (pull_request) Successful in 1m6s
Details

Browse Source 
PR #48 closed; treat the implementation as starting from
main, where no tmux integration exists yet. The PRD now
describes the full design (including the `_in_tmux` detection
+ helper scaffolding) as fresh work. Sized into 4 chunks:
`claude_docker_argv` refactor → tmux helpers + pane state +
`_attach_to_bottle` dispatch → new-agent flow → stop +
indicator.

Same design as before — opt-in by `\$TMUX`, split-window-then-
respawn, falls back to handoff on tmux failure or missing
binary. No external references to PR #48.
This commit is contained in:
didericis
2026-05-26 14:18:24 -04:00
parent 8b8d668602
commit e5316be454
1 changed files with 204 additions and 189 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/prds/0021-dashboard-tmux-split-pane.md
+204 -189
View File
@@ -10,55 +10,34 @@ When the dashboard runs inside tmux, lay it out as the **left
pane** of a two-pane window with the **selected agent's claude
session in the right pane**. Pressing Enter on an agent in the
agents pane swaps the right pane to that agent's session
(respawning with `claude --continue` so the conversation
picks up where it left off). Pressing `n` to start a new agent
spawns it directly into the right pane. The dashboard is the
(respawning with `claude --continue` so the conversation picks
up where it left off). Pressing `n` to start a new agent spawns
it directly into the right pane. The dashboard is the
operator's persistent left-hand surface; claude is always
visible to its right.
Outside tmux, fall back to today's handoff (`curses.endwin` →
foreground claude → `stdscr.refresh`). The split-pane UX is
Outside tmux, fall back to today's handoff (`curses.endwin →
foreground claude → stdscr.refresh`). The split-pane UX is
opt-in by environment — running the dashboard inside a tmux
session enables it, no flag required.
This supersedes the "new-window per agent" shape of PR #48
(option 3 from the research doc) with a tighter,
selection-driven model where the right pane reflects what the
operator is *currently focused on* rather than accumulating one
window per ever-attached bottle.
## Problem
PR #48 (and option 3 from `docs/research/claude-code-pane-in-
dashboard.md`) gets us partway there: when `$TMUX` is set, the
dashboard's Enter / `n` keys spawn claude in a new tmux
**window** (separate tab) instead of taking over the terminal.
That's better than the handoff in two respects — the dashboard
keeps refreshing, and multiple agents get their own panes — but
it has three real rough edges:
The dashboard's "Enter to attach" (PRD 0020 chunk 3) and its
new-agent `n` flow both take over the terminal for the
duration of the claude session via `curses.endwin`. While
claude has the screen, the dashboard's proposal queue and
agents pane are invisible. Any tool call that lands during a
claude session is silent until the operator exits back to the
dashboard.
1. **No simultaneous view.** A tmux window is a separate tab.
The operator switches to claude via `C-b n`, away from the
dashboard. Watching proposals queue while talking to claude
is back to the two-terminals-side-by-side workflow the
dashboard was supposed to collapse.
2. **Window accumulation.** Each agent ever attached adds a
tmux window. After a day with 10 bottles the operator has 10
windows to navigate; the dashboard's agents pane and tmux's
window list duplicate each other.
3. **Selection doesn't drive layout.** The dashboard's
selection model (PRD 0019) already knows which agent the
operator cares about right now. The new-window shape ignores
that — the operator has to manually switch tmux windows
independent of the dashboard's cursor position.
The split-pane shape this PRD proposes uses the same selection
the dashboard already tracks: cursor on agent X + Enter →
right pane shows agent X. Move cursor to agent Y + Enter →
right pane respawns with agent Y. One window, two panes, no
accumulation.
The split-pane shape this PRD proposes keeps both surfaces
visible: dashboard always on the left, the currently-selected
agent on the right. The dashboard's existing selection model
(PRD 0019) drives which agent occupies the right pane —
moving the cursor in the agents pane is a no-op; pressing
Enter swaps the right pane to that agent's session. One
window, two panes, no terminal handoff.
## Goals / Success Criteria
@@ -73,72 +52,73 @@ accumulation.
3. Pressing `n` to start a new agent (the existing chunk-2 flow
from PRD 0020) directs the spawned claude session into the
right pane instead of taking over the terminal.
4. Pressing `x` to stop a dashboard-owned bottle (PRD 0020
chunk 4): if that bottle was the right-pane occupant, the
right pane is cleared (or shows a brief "stopped" message).
4. Pressing `x` to stop a dashboard-owned bottle: if that
bottle was the right-pane occupant, the right pane is
cleared (or shows a brief "stopped" message).
5. Closing the right pane manually via tmux (e.g., `C-b x`)
leaves the dashboard intact; the next Enter creates a fresh
right pane.
6. Outside tmux (`$TMUX` unset), the dashboard's Enter / `n`
behavior falls back to today's handoff. No tmux dependency
for non-tmux users.
7. The two-pane layout works whether the operator pre-arranges
tmux themselves or lets the dashboard set it up. Both paths
land at the same place.
## Non-goals
- **The embedded-emulator option (option 2 from the research
doc).** This PRD stays on the multiplexer-delegated shape.
pyte-driven in-curses rendering is a separate, much larger
decision.
- **The embedded-emulator option** from
`docs/research/claude-code-pane-in-dashboard.md`. This PRD
stays on the multiplexer-delegated shape. pyte-driven
in-curses rendering is a separate, much larger decision.
- **Multi-pane / grid layout.** No "show 4 agents at once."
One left, one right. Picking that as a constraint dramatically
simplifies the state machine.
One left, one right. Picking that as a constraint
dramatically simplifies the state machine.
- **Persistence across dashboard exits.** The tmux session
state (which agent is in the right pane) doesn't survive a
dashboard restart. The bottles themselves persist (PRD 0020's
`q`-doesn't-tear-down), but reattaching is one Enter away
after a fresh launch.
- **Replacing PR #48 entirely while it's in flight.** This PRD
ships AFTER PR #48 lands (or merges over it). The shared
helpers (`_attach_via_tmux`, `_build_tmux_attach_argv`) carry
forward; the dispatch logic changes from `new-window` to
`split-window` / `respawn-pane`.
`q`-doesn't-tear-down); reattaching is one Enter away after a
fresh launch.
- **Cross-tmux-session orchestration.** The dashboard owns
panes in its own session. Other tmux sessions on the same
host are untouched.
- **A "right pane is detached" mode** where claude runs in a
pane that's not visible until expanded. Out of v1.
- **Auto-execing into a fresh tmux session** when launched
outside one. See open question #3.
## Scope
### In scope
- A `_tmux_split_pane_layout()` helper that, on first attach,
runs `tmux split-window -h …` to create the right pane and
remembers its tmux pane id so subsequent attaches use
`tmux respawn-pane -t <id> …` to swap content.
- Pane-id state on the main loop: `right_pane_id: str | None`.
None means "no right pane yet" → next attach creates one
via split-window. Non-None → respawn-pane.
- A `_pane_exists(pane_id)` check via `tmux list-panes`
before respawn, so a manually-closed right pane gracefully
falls back to a new split.
- Updated dispatch in `_attach_to_bottle` and `_new_agent_flow`:
- `$TMUX` set → split-pane / respawn-pane path
- `$TMUX` unset → existing handoff
- PR #48's `tmux new-window` path becomes the fallback inside
the split-pane code (e.g., if `tmux split-window` fails, or
via a config knob).
- `_stop_bottle_flow` clears `right_pane_id` if the stopped
bottle was the right-pane occupant.
- A `_in_tmux()` helper that returns True when `$TMUX` is set,
with `FileNotFoundError`-safe `subprocess.run` calls
around every tmux invocation so a missing tmux binary
cleanly falls back to handoff mode.
- Two new state fields on the main loop: `right_pane_id: str |
None` (the tmux pane id we created — None = "no right pane
yet, next attach must split") and `right_pane_slug: str |
None` (which bottle is currently in the right pane, for
the stop + indicator hooks).
- A `_tmux_split_pane_create(bottle, *, resume) -> str | None`
helper that runs `tmux split-window -h -P -F '#{pane_id}' …`
and returns the new pane id, or None on failure.
- A `_tmux_respawn_pane(pane_id, bottle, *, resume) -> bool`
helper that runs `tmux respawn-pane -k -t <id> …` and
returns success.
- A `_tmux_pane_exists(pane_id) -> bool` check via
`tmux list-panes -F '#{pane_id}'` so a manually-closed right
pane gracefully falls back to a new split.
- Modified `_attach_to_bottle` and `_new_agent_flow` (PRD
0020) to dispatch through the tmux helpers when `$TMUX` is
set, falling back to the existing handoff otherwise.
- `_stop_bottle_flow` updated to clear `right_pane_slug` when
the stopped bottle matches.
- An indicator in the agents pane (e.g., `*` prefix or
alternate attr) marking the row whose bottle is currently
in the right pane.
### Out of scope
- Choosing the split direction (-h vs -v). The PRD assumes
horizontal split (left/right). Open question if the operator
wants vertical.
- Choosing the split direction (-h vs -v) as a runtime knob.
v1 is `-h` (left / right). See open question #1.
- Sizing the split (e.g., 40/60 vs 50/50). v1 takes the tmux
default; sizing knob deferred.
- Persisting the right-pane occupant slug to disk so a fresh
@@ -148,134 +128,170 @@ accumulation.
### State machine
Two new fields on the main loop:
```python
# new state on _main_loop:
right_pane_id: str | None = None # tmux pane id we created
right_pane_slug: str | None = None # which bottle's session is in it
right_pane_id: str | None = None # tmux pane id, or None
right_pane_slug: str | None = None # current occupant, or None
```
Dispatch logic:
### Dispatch
`_attach_to_bottle` and `_new_agent_flow` (from PRD 0020) gain
a tmux branch:
```python
def _spawn_into_right_pane(bottle, slug, *, resume) -> str:
if not _in_tmux():
# Fall back to today's handoff.
return _attach_via_handoff(stdscr, bottle, slug, resume=resume)
def _attach_in_tmux(bottle, slug, *, resume) -> str:
nonlocal right_pane_id, right_pane_slug
if right_pane_id and _pane_exists(right_pane_id):
# Respawn content in the existing pane.
ok = _tmux_respawn_pane(right_pane_id, bottle, resume=resume)
if not ok:
right_pane_id = None # fall through to create
if right_pane_id is None or not _pane_exists(right_pane_id):
right_pane_id = _tmux_split_pane_create(bottle, resume=resume)
if right_pane_id is None:
# tmux failed; fall back to handoff.
return _attach_via_handoff(stdscr, bottle, slug, resume=resume)
# If we have a remembered pane and it still exists, respawn it.
if right_pane_id and _tmux_pane_exists(right_pane_id):
if _tmux_respawn_pane(right_pane_id, bottle, resume=resume):
right_pane_slug = slug
return f"[{slug}] in right pane"
right_pane_id = None # respawn failed; fall through
# Otherwise create a fresh split.
pane_id = _tmux_split_pane_create(bottle, resume=resume)
if pane_id is None:
# tmux failed; fall back to handoff.
return _attach_via_handoff(stdscr, bottle, slug, resume=resume)
right_pane_id = pane_id
right_pane_slug = slug
return f"[{slug}] in right pane"
```
### Pane creation + respawn
The non-tmux path is unchanged from PRD 0020 — `_attach_via_
handoff` is what those two flows already do today (curses.
endwin → attach_claude → stdscr.refresh).
### Pane creation
`tmux split-window -h -P -F '#{pane_id}'` opens a right pane
and prints its pane id (e.g., `%12`). The `-h` is horizontal
split (left/right); the `-P -F` combo emits the new pane's
identifier for tracking.
```python
def _tmux_split_pane_create(bottle, *, resume) -> str | None:
"""Open a right pane via `tmux split-window -h -P -F '#{pane_id}'`.
The -P -F combo prints the new pane's id to stdout so we
can track it for respawn."""
docker_argv = bottle.claude_docker_argv(_claude_args(resume=resume))
result = subprocess.run(
["tmux", "split-window", "-h",
"-P", "-F", "#{pane_id}",
*docker_argv],
capture_output=True, text=True, check=False,
)
try:
result = subprocess.run(
["tmux", "split-window", "-h",
"-P", "-F", "#{pane_id}",
*docker_argv],
capture_output=True, text=True, check=False,
)
except FileNotFoundError:
return None
if result.returncode != 0:
return None
return result.stdout.strip()
return result.stdout.strip() or None
```
### Pane respawn
`tmux respawn-pane -k -t <id> …` replaces the pane's content
without re-splitting. `-k` kills the existing process first
(claude session ends; the underlying bottle is untouched).
`--continue` on the new claude invocation picks up the prior
conversation.
```python
def _tmux_respawn_pane(pane_id, bottle, *, resume) -> bool:
"""Replace the pane's content via `tmux respawn-pane -t <id> -k …`.
-k kills the existing process before respawning. Idempotent
on a pane that's already empty."""
docker_argv = bottle.claude_docker_argv(_claude_args(resume=resume))
result = subprocess.run(
["tmux", "respawn-pane", "-k", "-t", pane_id, *docker_argv],
capture_output=True, text=True, check=False,
)
try:
result = subprocess.run(
["tmux", "respawn-pane", "-k", "-t", pane_id, *docker_argv],
capture_output=True, text=True, check=False,
)
except FileNotFoundError:
return False
return result.returncode == 0
```
### Pane-existence check
`tmux list-panes -F '#{pane_id}'` lists pane ids in the current
window. The dashboard checks for `right_pane_id` membership
before issuing respawn-pane; if absent (operator closed the
pane via `C-b x`), it falls back to creating a fresh split.
```python
def _tmux_pane_exists(pane_id) -> bool:
try:
result = subprocess.run(
["tmux", "list-panes", "-F", "#{pane_id}"],
capture_output=True, text=True, check=False,
)
except FileNotFoundError:
return False
if result.returncode != 0:
return False
return pane_id in result.stdout.splitlines()
```
### Bottle.claude_docker_argv
The tmux helpers need the full docker-exec argv for claude —
specifically including the `--append-system-prompt-file <path>`
flag that `DockerBottle.exec_claude` appends today when the
bottle has a prompt path. Refactor: split `exec_claude` into a
pure `claude_docker_argv(args, *, tty)` that returns the argv
and a thin `exec_claude` that calls `subprocess.run` on it.
Both the tmux path AND the existing foreground path use the
same argv builder.
### Stop integration
`_stop_bottle_flow` already pops the bottle from the
`bottles` dict and calls `cm.__exit__`. The new bit: if the
stopped slug matches `right_pane_slug`, clear both pane
tracking variables. The tmux pane itself stays open with
docker exec's "container not found" error showing — acceptable;
the operator hits Enter on a different agent to repurpose it
(respawn-pane will replace the broken state).
`_stop_bottle_flow` (from PRD 0020 chunk 4) gets a small
addition: after the bottle is torn down, if `right_pane_slug
== stopped_slug`, clear it. The tmux pane itself stays open
with claude's "container not found" error showing. The next
Enter on a different agent will respawn it.
### Indicator in agents pane
When `right_pane_slug` matches an agent row's slug, render
that row with a `*` prefix (and / or reverse-video, if the
focus indicator doesn't already conflict). Gives the operator
a clear visual link between dashboard selection and right-
pane content.
### Fallback when tmux fails
Three failure modes worth handling:
1. **`$TMUX` set but tmux binary not on PATH** (rare but
possible in nested containers). Detected when
`subprocess.run` raises `FileNotFoundError` — treat as not-
in-tmux, fall back to handoff.
2. **`tmux split-window` returns non-zero** (e.g., pane size
too small, server lost). Surface a status-line error and
fall back to handoff for that one keypress.
1. **`$TMUX` set but tmux binary not on PATH.** `subprocess.run`
raises `FileNotFoundError` — treat as not-in-tmux, fall back
to handoff. Should be rare but happens in nested
containers.
2. **`tmux split-window` / `respawn-pane` returns non-zero.**
Surface a status-line error and fall back to handoff for
that one keypress. The dashboard stays usable.
3. **Right pane manually closed.** Detected at next attach via
pane-exists check; create fresh split.
### Interaction with PR #48
PR #48 introduced `_in_tmux()`, `_attach_via_tmux` (using
`tmux new-window`), and `_build_tmux_attach_argv`. This PRD
reuses the first; rewrites the second; the third (window
argv builder) becomes one of several tmux-argv builders this
PRD introduces. Net change: PR #48's new-window behavior is
replaced by split-pane behavior. The diff is moderate but
contained — the abstraction shape carries through.
`_tmux_pane_exists`; create a fresh split.
## Implementation chunks
Sized small.
1. **Pane state + create path.** Add `right_pane_id` /
`right_pane_slug` to the main loop; implement
`_tmux_split_pane_create`; wire Enter / `n` to use it when
`right_pane_id is None`. No respawn yet — second attach
creates a *second* split-pane (visibly wrong, but isolated).
2. **Respawn for subsequent attaches.** Implement
`_tmux_respawn_pane` + `_pane_exists`; route second-and-
subsequent attaches through respawn instead of split. The
"wrong" behavior from chunk 1 disappears.
3. **Stop integration.** `_stop_bottle_flow` clears
`right_pane_slug` when the stopped bottle matches; tests.
4. **Replace PR #48's new-window with split-pane.** Remove the
`tmux new-window` invocation; keep `_build_tmux_attach_argv`
as a shared helper if useful. Update PR #48's tests to
match the new behavior. (If PR #48 has merged by then, this
is a follow-up commit; if not, the two PRs reconcile at
merge time.)
1. **`claude_docker_argv` refactor.** Pure-ish split of
`DockerBottle.exec_claude` so both foreground and tmux
paths build on the same argv. No behavior change for the
existing tests.
2. **tmux helpers + pane state.** Add `_in_tmux`,
`_tmux_split_pane_create`, `_tmux_respawn_pane`,
`_tmux_pane_exists`, plus the two new state fields on the
main loop. Wire `_attach_to_bottle` to dispatch through
tmux when in tmux, fall back to handoff otherwise.
3. **New-agent flow integration.** `_new_agent_flow` (PRD
0020 chunk 2) gains the same tmux dispatch — first attach
on a freshly-started agent goes into the right pane.
4. **Stop integration + right-pane indicator.**
`_stop_bottle_flow` clears `right_pane_slug` when matched;
`_format_agent_row` (PRD 0019 chunk 2) gets a `*` prefix
when its slug matches `right_pane_slug`.
## Open questions
1. **Split direction: horizontal (`-h`) vs vertical (`-v`).**
The PRD's "left / right" framing implies `-h`. But on a
The PRD's "left / right" framing implies `-h`. On a
short-and-wide terminal a vertical (top / bottom) split
might give the dashboard more vertical real estate. Pick
`-h` for v1; revisit if the dashboard's agents-pane row
@@ -292,10 +308,10 @@ Sized small.
3. **Dashboard launched OUTSIDE tmux but tmux is installed.**
Should the dashboard auto-exec itself inside a fresh tmux
session to get the split-pane experience? Convenient but
surprising (`./cli.py dashboard` shouldn't silently change
what session you're in). v1 leaves this off — operators
who want split-pane mode start tmux themselves and then
run the dashboard.
surprising (`./cli.py dashboard` shouldn't silently
change what session you're in). v1 leaves this off —
operators who want split-pane mode start tmux themselves
and then run the dashboard.
4. **Pane size after split.** tmux's default for `split-window
-h` is 50/50. The dashboard's column widths are designed
@@ -304,26 +320,26 @@ Sized small.
want ~50 cols and claude gets the rest. Add `-p <pct>` or
`-l <cols>` flag? Open question; v1 uses tmux's default.
5. **Multi-window tmux sessions.** If the operator has the
dashboard in tmux window A and unrelated work in window B,
the split-pane goes into window A as expected (tmux
commands are window-scoped by `$TMUX_PANE` context). No
special handling needed — but worth verifying behavior
when the operator switches away from window A before the
dashboard issues a split.
5. **Cursor highlighting in the agents pane to indicate "this
one's in the right pane right now."** Resolved YES in
scope — `*` prefix on the matching row. Open: should it
also use color, and which color doesn't conflict with the
PRD-0019 focus indicator?
6. **Cursor highlighting in the agents pane to indicate "this
one's in the right pane right now."** The dashboard could
mark the slug currently occupying the right pane with a
`*` prefix or alternate color in the agents pane. Probably
yes for v1 — operator clarity. The state's already
tracked (`right_pane_slug`).
6. **Concurrent dashboards in different tmux windows.**
Multiple `./cli.py dashboard` invocations in different
tmux windows would each create their own right pane —
probably fine, each has its own state, but worth
verifying that `tmux list-panes` is scoped to the right
window context.
7. **What about the proposals pane?** Approvals (`a`/`m`/`r`)
are unchanged — they don't open new panes, they apply
in-process. The proposals pane stays exactly as today,
just narrower because it shares the left pane with the
agents list.
7. **`$TMUX_PANE` vs `$TMUX`.** Tmux sets both. We only check
`$TMUX` (presence of the server socket); `$TMUX_PANE` is
the dashboard's own pane id, which we could capture to be
more precise about `tmux split-window -t <pane>` (split
relative to the dashboard's pane specifically). Probably
not needed since `split-window` defaults to the current
active pane, but worth verifying.
## References
@@ -333,7 +349,6 @@ Sized small.
(`_attach_to_bottle`, `_new_agent_flow`, `_stop_bottle_flow`
— the three hooks this PRD changes)
- `docs/research/claude-code-pane-in-dashboard.md` — the
three-option survey; this PRD picks option 3 with a tighter
three-option survey of how to surface claude inside the
dashboard; this PRD picks option 3 with a tighter
split-pane variant
- PR #48 — opt-in tmux via `tmux new-window`; this PRD
supersedes its window-mode with pane-mode