hexcli 2.19.0__tar.gz → 2.20.0__tar.gz

{hexcli-2.19.0 → hexcli-2.20.0}/CHANGELOG.md

@@ -4,6 +4,37 @@ Full evidence for every claim below — including the experiments that failed
 lives in `docs/V2_PLAN.md` §14. Numbers are pass^k over repeated live runs on
 the Hexagon NPU, not single-run anecdotes.
 
+## 2.20.0 — 2026-09-18
+
+A minor release: a tool result the model reads changes (`run_code` on a
+program that waits for the keyboard).
+
+> **Gate: PASS** on the pinned 24-case set: 21 cases held 5/5 in the arm;
+> `casual-4`, `factual-4` and `memory-1` lost four, two and four runs to
+> backend timeouts and were re-run on a fresh server at six runs, 6/6 each.
+> Own arm, seed 20260918. **Platform: 13 invalid of 245 runs (5.3 %)**, the
+> server's working set again above 6 GB with 1.6 GB free, so the ceiling
+> comparison carries the `[PLATFORM]` flag: run-level 402/514 vs 166/217,
+> −1.7 %, Fisher p = 0.63; pass^k 26 → 30 of 46 shared cases, McNemar
+> p = 0.13. The change did what it is for: `run_code` timed out 0 times in
+> the arm against 10 in the 2.12.0 arm (all of them `make-py-1`'s
+> calculator waiting on `input()`), the note fired twice, and `make-py-1`
+> stayed 5/5. Smoke 10/10 on a fresh server.
+
+- `run_code` gives the program no keyboard, and says so when the program
+  asked for one. The child inherited the terminal's standard input, so a
+  program that calls `input()` sat waiting on the owner's console until the
+  tool's timeout: on 2026-09-18 "create a python cli hilo guessing game and
+  run it" wrote a working game and then waited the full ten seconds, twice,
+  to report "TIMEOUT". Standard input is now closed for the child, so
+  `input()` raises `EOFError` at once, and when it does the result adds one
+  line: the program reads from the keyboard, `run_code` has none to give
+  it, and it has to be started in a terminal, with the command. On record:
+  12 `run_code` timeouts whose output ends in a prompt (ten of them the
+  `make-py-1` calculator) and 19 runs that already hit `EOFError` under the
+  eval runner, which has no terminal; 0.6 % of recorded runs, all of them
+  this shape.
+
 ## 2.19.0 — 2026-09-18
 
 A minor release: nudge text the model reads is new (the claimed-fix rule)

{hexcli-2.19.0 → hexcli-2.20.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: hexcli
-Version: 2.19.0
+Version: 2.20.0
 Summary: Local Hexagon NPU terminal agent for Snapdragon X Elite Windows ARM64
 Project-URL: Homepage, https://github.com/NathanL15/Hex-CLI
 Project-URL: Repository, https://github.com/NathanL15/Hex-CLI

{hexcli-2.19.0 → hexcli-2.20.0}/hexcli/__init__.py

@@ -3,4 +3,4 @@
 # The one place the version is written. pyproject.toml reads it (hatch
 # dynamic version), agent.VERSION re-exports it, and CI refuses a release
 # tag that does not match it.
-__version__ = "2.19.0"
+__version__ = "2.20.0"

{hexcli-2.19.0 → hexcli-2.20.0}/hexcli/tools.py

@@ -918,8 +918,13 @@ def run_code_tool(
             )
     cmd = [*cmd_prefix, str(path), *[str(a) for a in run_args]]
     try:
+        # No keyboard. With the parent's stdin inherited, a program that
+        # calls input() sat on the owner's terminal until the timeout (10 s,
+        # twice, 2026-09-18: "create a python cli hilo guessing game and run
+        # it"); with stdin closed it raises EOFError at once, and the note
+        # below tells the model what that means.
         proc = subprocess.Popen(
-            cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE,
+            cmd, stdin=subprocess.DEVNULL, stdout=subprocess.PIPE, stderr=subprocess.PIPE,
             text=True, encoding="utf-8", errors="replace",
         )
     except (PermissionError, FileNotFoundError, OSError) as exc:
@@ -940,10 +945,16 @@ def run_code_tool(
     out = trim_text(stdout, output_limit)
     err = trim_text(stderr, output_limit)
     ui.tool_event("run", f"{path}  (exit {proc.returncode})")
+    note = ""
+    if "EOFError" in stderr:
+        note = (f"\n\nThe program stopped where it reads from the keyboard (input()). "
+                f"run_code has no keyboard to give it, so it cannot be played or driven "
+                f"here; it has to be started in a terminal: python {path.name}. The code "
+                f"up to that point ran.")
     return (
         f"Exit code: {proc.returncode}\n\n"
         f"[stdout]\n{out or '(empty)'}\n\n"
-        f"[stderr]\n{err or '(empty)'}"
+        f"[stderr]\n{err or '(empty)'}{note}"
     )