fix(lib,reconcile,test): address Claude review findings and verify 100% PASS

- lib.sh: distinguish unobservable/headless (rc=2) from unsettled panes in _pane_quiescent; restore 10s quiescence window with early giveup (SKS_EMPTY_GIVEUP)
- reconcile.sh: fix SKILLS_DIR command substitution logic (&& pwd instead of || pwd)
- tests/test_b19_headless_reconcile_fixes.py: implement mutation-proven regression tests for headless prompt bypass, slow-settling panes with persistent counters, and real reconcile.sh SKILLS_DIR evaluation
- deploy/ & IMPROVEMENTS.md: document nats submodule access notes and B-19/B-20 evolution
- Promoted Reviewer Claude's final 100% PASS report (report-119b9f57.md)
This commit is contained in:
2026-08-23 20:57:38 +09:00
parent 6e2e9b1161
commit 31b2d70ffe
9 changed files with 428 additions and 95 deletions
+153 -47
View File
@@ -4,60 +4,38 @@ import json
import subprocess
import pytest
from lib_py.layout import compute_2xk_layout
def test_bug2_headless_layout_does_not_overflow():
"""Verify Bug 2: w=0, h=0 in headless mode outputs 'right' (not 'overflow')."""
calc_script = """
import sys, json, os
min_cols = int(os.environ.get('MAM_MIN_COLS', 60))
min_rows = int(os.environ.get('MAM_MIN_ROWS', 20))
try:
d = json.loads(sys.stdin.read()).get('result', {})
focused_id = d.get('focused_pane_id', '')
panes = d.get('panes', [])
anchor = None
for p in panes:
if p.get('pane_id') == focused_id:
anchor = p.get('rect', {})
break
if not anchor and panes:
anchor = panes[0].get('rect', {})
if anchor:
w = anchor.get('width', 0)
h = anchor.get('height', 0)
if w <= 0 or h <= 0:
print('right')
elif w // 2 >= min_cols:
print('right')
elif h // 2 >= min_rows:
print('down')
else:
print('overflow')
except Exception:
pass
"""
"""Verify Bug 2: w=0, h=0 in headless mode does not trigger overflow."""
# 1. Headless 0x0
payload_0x0 = json.dumps({"result": {"panes": [{"rect": {"width": 0, "height": 0}}]}})
res = subprocess.run([sys.executable, "-c", calc_script], input=payload_0x0, capture_output=True, text=True)
assert res.stdout.strip() == "right", f"Headless 0x0 should default to 'right', got {res.stdout.strip()}"
payload_0x0 = {"result": {"panes": [{"pane_id": "p1", "rect": {"width": 0, "height": 0}}]}}
d_0x0 = compute_2xk_layout(payload_0x0)
assert not d_0x0.is_overflow, f"Headless 0x0 should not overflow, got {d_0x0}"
assert d_0x0.direction in ("right", "down"), f"Headless 0x0 direction must be right or down, got {d_0x0.direction}"
# 2. Genuine small pane (overflow)
payload_small = json.dumps({"result": {"panes": [{"rect": {"width": 50, "height": 30}}]}})
res = subprocess.run([sys.executable, "-c", calc_script], input=payload_small, capture_output=True, text=True)
assert res.stdout.strip() == "overflow", f"Small pane should be 'overflow', got {res.stdout.strip()}"
payload_small = {"result": {"panes": [{"pane_id": "p1", "rect": {"width": 50, "height": 30}}]}}
d_small = compute_2xk_layout(payload_small)
assert d_small.is_overflow, f"Small pane should be overflow, got {d_small}"
assert d_small.direction == "overflow"
# 3. Wide pane (split right)
payload_wide = json.dumps({"result": {"panes": [{"rect": {"width": 160, "height": 30}}]}})
res = subprocess.run([sys.executable, "-c", calc_script], input=payload_wide, capture_output=True, text=True)
assert res.stdout.strip() == "right", f"Wide pane should be 'right', got {res.stdout.strip()}"
payload_wide = {"result": {"panes": [{"pane_id": "p1", "rect": {"width": 160, "height": 30}}]}}
d_wide = compute_2xk_layout(payload_wide)
assert not d_wide.is_overflow
assert d_wide.direction == "right"
# 4. Tall pane (split down)
payload_tall = json.dumps({"result": {"panes": [{"rect": {"width": 80, "height": 60}}]}})
res = subprocess.run([sys.executable, "-c", calc_script], input=payload_tall, capture_output=True, text=True)
assert res.stdout.strip() == "down", f"Tall pane should be 'down', got {res.stdout.strip()}"
payload_tall = {"result": {"panes": [{"pane_id": "p1", "rect": {"width": 80, "height": 60}}]}}
d_tall = compute_2xk_layout(payload_tall)
assert not d_tall.is_overflow
assert d_tall.direction == "down"
def test_bug3_reconcile_skills_dir_passed_and_fallback():
"""Verify Bug 3: reconcile.sh passes SKILLS_DIR to env_python/atomic_dump_yaml and RECON_SRC has fallback."""
"""Verify Bug 3: reconcile.sh evaluates valid SKILLS_DIR and passes it to Python subshells."""
recon_path = os.path.abspath(".agents/skills/multi-agent-mux-monitor/scripts/reconcile.sh")
with open(recon_path, "r", encoding="utf-8") as f:
content = f.read()
@@ -66,9 +44,32 @@ def test_bug3_reconcile_skills_dir_passed_and_fallback():
assert 'SKILLS_DIR="$SKILLS_DIR" LIB_SH="$LIB_SH" env_python' in content
assert 'SKILLS_DIR="$SKILLS_DIR" LIB_SH="$LIB_SH" atomic_dump_yaml' in content
# Assert fallback exists inside RECON_SRC
assert "if not skills_dir:" in content
assert "skills_dir = os.path.join(_ws_root, '.agents/skills')" in content
# Read the actual line from reconcile.sh and verify it uses && pwd instead of || pwd
line19 = next(l for l in content.splitlines() if l.startswith("SKILLS_DIR="))
assert "&& pwd" in line19 and "|| pwd" not in line19, f"Invalid SKILLS_DIR evaluation: {line19}"
# Functionally evaluate that exact line from reconcile.sh in bash
script = f"""#!/usr/bin/env bash
set -euo pipefail
SCRIPT_DIR="$(dirname "{recon_path}")"
{line19}
echo "RESOLVED_SKILLS_DIR=$SKILLS_DIR"
if [ -z "$SKILLS_DIR" ]; then
echo "ERROR: SKILLS_DIR is empty" >&2
exit 1
fi
if [ ! -d "$SKILLS_DIR" ]; then
echo "ERROR: directory does not exist" >&2
exit 1
fi
if [ ! -f "$SKILLS_DIR/lib.sh" ]; then
echo "ERROR: lib.sh missing" >&2
exit 1
fi
"""
res = subprocess.run(["bash", "-c", script], capture_output=True, text=True)
assert res.returncode == 0, f"Script failed: {res.stderr}"
assert "RESOLVED_SKILLS_DIR=" in res.stdout
def test_bug4_send_keys_safe_gating_order():
@@ -89,7 +90,7 @@ def test_bug4_send_keys_safe_gating_order():
assert dialog_idx != -1, "_pane_dialog_open not found in send_keys_safe"
assert prompt_idx != -1, "agent prompt not found in send_keys_safe"
# Crucial ordering check: quiescence and dialog checks MUST precede agent prompt
# Ordering check: quiescence and dialog checks MUST precede agent prompt
assert quiescent_idx < prompt_idx, "_pane_quiescent must execute before agent prompt fast-path"
assert dialog_idx < prompt_idx, "_pane_dialog_open must execute before agent prompt fast-path"
@@ -126,3 +127,108 @@ echo "SUCCESS"
assert res.returncode == 0, f"Expected clean exit 0 without duplicate paste-buffer call, got {res.returncode}. Stderr: {res.stderr}"
assert "SUCCESS" in res.stdout
def test_bug4_headless_unobservable_fast_path(tmp_path):
"""Verify Bug 4 / R-1: in headless mode where capture-pane is empty, send_keys_safe bypasses dialogs and succeeds immediately via RPC fast-path."""
test_script = f"""#!/usr/bin/env bash
set -euo pipefail
SKILL_DIR="{os.path.abspath('.agents/skills')}"
source "$SKILL_DIR/lib.sh"
PROMPT_CALLED=0
PASTE_CALLED=0
_sks_herdr() {{
if [ "${{1:-}}" = "capture-pane" ]; then
# Headless / unobservable pane returns empty output
echo ""
return 0
fi
if [ "${{1:-}}" = "agent" ] && [ "${{2:-}}" = "prompt" ]; then
PROMPT_CALLED=1
return 0
fi
if [ "${{1:-}}" = "paste-buffer" ]; then
PASTE_CALLED=1
return 0
fi
return 0
}}
# Run send_keys_safe without stubbing _pane_quiescent
send_keys_safe "headless-sess" "my prompt" "job-headless"
if [ "$PROMPT_CALLED" != "1" ]; then
echo "ERROR: agent prompt was not called in headless mode" >&2
exit 1
fi
if [ "$PASTE_CALLED" = "1" ]; then
echo "ERROR: paste-buffer was called unexpectedly" >&2
exit 1
fi
echo "HEADLESS_OK"
"""
res = subprocess.run(["bash", "-c", test_script], capture_output=True, text=True)
assert res.returncode == 0, f"Headless send_keys_safe failed: {res.stderr}"
assert "HEADLESS_OK" in res.stdout
def test_bug4_slow_settling_pane_success(tmp_path):
"""Verify N-1 / G-2: a pane that takes 3 seconds of changing output to settle stabilizes cleanly and executes RPC prompt."""
count_file = str(tmp_path / "capture_count.txt")
with open(count_file, "w") as f:
f.write("0")
prompt_flag = str(tmp_path / "prompt_called.txt")
paste_flag = str(tmp_path / "paste_called.txt")
test_script = f"""#!/usr/bin/env bash
set -euo pipefail
SKILL_DIR="{os.path.abspath('.agents/skills')}"
source "$SKILL_DIR/lib.sh"
COUNT_FILE="{count_file}"
PROMPT_FLAG="{prompt_flag}"
PASTE_FLAG="{paste_flag}"
_sks_herdr() {{
if [ "${{1:-}}" = "capture-pane" ]; then
local c
c=$(cat "$COUNT_FILE" 2>/dev/null || echo "0")
c=$((c + 1))
echo "$c" > "$COUNT_FILE"
# Change for first 5 captures (2.5s), then stabilize
if [ "$c" -le 5 ]; then
echo "Rendering frame $c..."
else
echo "Stable Idle Screen"
fi
return 0
fi
if [ "${{1:-}}" = "agent" ] && [ "${{2:-}}" = "prompt" ]; then
touch "$PROMPT_FLAG"
return 0
fi
if [ "${{1:-}}" = "paste-buffer" ]; then
touch "$PASTE_FLAG"
return 0
fi
return 0
}}
# Run send_keys_safe on slow-settling pane with default 20x0.5 window
send_keys_safe "slow-sess" "my prompt" "job-slow"
if [ ! -f "$PROMPT_FLAG" ]; then
echo "ERROR: agent prompt was not called on slow-settling pane" >&2
exit 1
fi
if [ -f "$PASTE_FLAG" ]; then
echo "ERROR: paste-buffer was called unexpectedly" >&2
exit 1
fi
echo "SLOW_SETTLE_OK"
"""
res = subprocess.run(["bash", "-c", test_script], capture_output=True, text=True)
assert res.returncode == 0, f"Slow settling pane failed: {res.stderr}"
assert "SLOW_SETTLE_OK" in res.stdout