fix(source-control): refuse flag abbreviation on every babysit entry point

plugins/source-control/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://json.schemastore.org/claude-code-plugin-manifest.json",
   "name": "source-control",
-  "version": "0.30.0",
+  "version": "0.31.0",
   "description": "Git and GitHub delivery workflow: /commit (Conventional Commits + Co-Authored-By trailer via safe heredoc mechanics), /pull-request (prep, create, CI monitoring, review-comment triage, merge, CI-log fetch), /babysit-prs (self-pacing fleet loop — safe by default; opt-in worker/autopilot tiers add gate-checked merge and thread resolution behind a deterministic Python engine), /babysit-loop (the loop-lane merge lane: a standing or drain loop that invokes babysit-prs per cycle, configured through repo-scoped babysit_loop_* keys on the layered source-control.md seam, with merge authority human-only until the target repo's tracked config adopts the lane, a gate-proven C2-mechanical baseline once adopted, and merge-rung raises binding from the team-tracked layer only), /worktree (create, status, cleanup, audit for parallel-session isolation), /setup (check the effective commit-subject / PR-title convention merged across its config layers and the babysit-prs config, or apply — interview the repo and write the convention config to a chosen layer), and /resolve-conflicts (intent-first merge/rebase conflict resolution with a semantic-conflict sweep — never --abort). The commit-subject / PR-title convention is configurable via a source-control.md config written by a re-runnable setup skill, layered across a ~/.claude user-global file, the tracked team file, and a gitignored .claude/source-control.local.md personal overlay merged per key; Conventional Commits is the default when no convention is declared.",
   "author": {
     "name": "Melodic Software",

plugins/source-control/CHANGELOG.md

@@ -3,6 +3,34 @@
 All notable changes to the `source-control` plugin are documented here. Format follows
 [Keep a Changelog](https://keepachangelog.com/en/1.1.0/); this plugin uses semantic versioning.
 
+## [0.31.0]
+
+### Changed
+
+- **No babysit parser resolves a flag abbreviation any more, and the property is now the
+  directory's rather than two files' (`#1371`).** A permission grant states its condition as the
+  literal presence or absence of a flag in the command text — above all "no `--merge` means
+  check-only". Argparse's default prefix abbreviation lets `--mer` resolve to `--merge` while the
+  command text contains no such flag, so the written command and the resolved behavior diverge,
+  which is exactly what such a condition must be able to rule out. `#1354` closed this on
+  `babysit_merge.py` and `babysit_resolve_thread.py`; the remaining seven entry points —
+  `babysit_findings.py`, `manage_babysit_lease.py`, `manage_feedback_ledger.py`,
+  `pr_queue_snapshot.py`, `prune_babysit_worktrees.py`, `refresh_pr_branch.py`, and
+  `request_review.py` — still inherited the default. All nine now set `allow_abbrev=False`.
+
+  Hardening them one at a time is what let the gap persist, so the guard contract gains a gate over
+  the whole catalogue: every Python entry point is invoked with an unambiguous three-character
+  prefix of `--help` and must not exit 0. `--help` is registered on every parser, and it
+  short-circuits parsing — so an abbreviation that resolves exits 0 before required-argument
+  validation runs, while one that does not is a usage error. That makes the exit code a sufficient
+  discriminator without a per-CLI argument shape, and a companion test asserts the discrimination
+  against argparse itself rather than assuming it. Three characters because
+  `manage_babysit_lease.py` also registers `--heartbeat-interval-seconds`, so a shorter prefix is
+  ambiguous there and exits 2 regardless — the probe would have passed on that entry point while
+  proving nothing. A tenth entry point arriving with the default now fails CI instead of shipping.
+
+  Abbreviated invocations that previously worked are now usage errors, which is the point.
+
 ## [0.30.0]
 
 ### Added

plugins/source-control/skills/babysit-prs/scripts/babysit_findings.py

@@ -134,7 +134,7 @@ def fetch_live_comments(repo: str, number: int) -> list[dict[str, Any]]:
 
 
 def parse_args(argv: list[str]) -> argparse.Namespace:
-    parser = argparse.ArgumentParser(add_help=True)
+    parser = argparse.ArgumentParser(add_help=True, allow_abbrev=False)
     source = parser.add_mutually_exclusive_group(required=True)
     source.add_argument("--pr", type=int, help="live PR number to fetch and count")
     source.add_argument("--comments-json", help="comments JSON file (no network)")

plugins/source-control/skills/babysit-prs/scripts/manage_babysit_lease.py

@@ -62,7 +62,7 @@ def run(args: argparse.Namespace) -> dict:
 
 def main() -> int:
     configure_stdio()
-    parser = argparse.ArgumentParser(description=__doc__)
+    parser = argparse.ArgumentParser(description=__doc__, allow_abbrev=False)
     parser.add_argument("action", choices=("acquire", "heartbeat", "release", "reap"))
     parser.add_argument(
         "--scope",

plugins/source-control/skills/babysit-prs/scripts/manage_feedback_ledger.py

@@ -173,7 +173,7 @@ def run_locked(
 
 def main() -> int:
     configure_stdio()
-    parser = argparse.ArgumentParser(description=__doc__)
+    parser = argparse.ArgumentParser(description=__doc__, allow_abbrev=False)
     parser.add_argument(
         "action",
         choices=("dispose", "record-advisory-round", "record-worker-checkin"),

plugins/source-control/skills/babysit-prs/scripts/pr_queue_snapshot.py

@@ -410,7 +410,7 @@ def exit_code_for(snapshot: dict[str, Any]) -> int:
 def main() -> int:
     configure_stdio()
     parser = argparse.ArgumentParser(
-        description="Read-only GitHub PR babysitting snapshot."
+        description="Read-only GitHub PR babysitting snapshot.", allow_abbrev=False
     )
     group = parser.add_mutually_exclusive_group(required=True)
     group.add_argument(

plugins/source-control/skills/babysit-prs/scripts/prune_babysit_worktrees.py

@@ -157,7 +157,9 @@ def active_worker_lease(
 
 def main() -> int:
     configure_stdio()
-    parser = argparse.ArgumentParser(description="Prune babysit-prs Git worktrees.")
+    parser = argparse.ArgumentParser(
+        description="Prune babysit-prs Git worktrees.", allow_abbrev=False
+    )
     parser.add_argument(
         "--root",
         required=True,

plugins/source-control/skills/babysit-prs/scripts/refresh_pr_branch.py

@@ -198,7 +198,7 @@ def run_locked(
 
 def main() -> int:
     configure_stdio()
-    parser = argparse.ArgumentParser(description=__doc__)
+    parser = argparse.ArgumentParser(description=__doc__, allow_abbrev=False)
     parser.add_argument("--pr", required=True, help="PR URL or owner/repo#number")
     parser.add_argument(
         "--expected-head-sha",

plugins/source-control/skills/babysit-prs/scripts/request_review.py

@@ -590,7 +590,7 @@ def run_locked(
 
 def main() -> int:
     configure_stdio()
-    parser = argparse.ArgumentParser(description=__doc__)
+    parser = argparse.ArgumentParser(description=__doc__, allow_abbrev=False)
     parser.add_argument("--pr", required=True, help="PR URL or owner/repo#number")
     parser.add_argument(
         "--expected-head-sha",

plugins/source-control/skills/babysit-prs/scripts/tests/test_guards.py

@@ -12,6 +12,9 @@
 consumers cite; `GeneratedDocIsCurrent` is what keeps it honest.
 """
 
+import argparse
+import contextlib
+import io
 import json
 import os
 import pathlib
@@ -359,6 +362,79 @@ def test_every_mechanism_row(self) -> None:
                     )
 
 
+class NoParserResolvesAnAbbreviation(unittest.TestCase):
+    """`allow_abbrev=False` on every catalogued Python entry point, not two of them.
+
+    A permission grant states its condition as the literal presence or absence of
+    a flag in the command text. Argparse's default prefix abbreviation lets
+    `--app` resolve to `--apply` while the text contains no such flag, so the
+    written command and the resolved behavior diverge -- which is precisely what
+    such a condition has to be able to rule out. The merge gate and the thread
+    resolver were hardened individually; the property belongs to the directory,
+    because the next entry point inherits the default unless something fails.
+
+    ABBREVIATION_PROBE is the universal one: argparse registers `--help` on every
+    parser here, so with abbreviation on it resolves and exits 0, and with
+    abbreviation off it is an unrecognized argument at exit 2. No per-CLI
+    argument shape is needed, which is what makes this a gate over the catalogue
+    rather than a hand-maintained list of cases.
+
+    Three characters, not one or two: `manage_babysit_lease.py` also registers
+    `--heartbeat-interval-seconds`, so a shorter prefix is AMBIGUOUS there and
+    exits 2 even with abbreviation on -- the probe would pass on that entry point
+    while proving nothing.
+    """
+
+    ABBREVIATION_PROBE = "--hel"  # spellchecker:disable-line
+
+    def test_every_python_entry_point_refuses_an_abbreviated_flag(self) -> None:
+        catalogued = [
+            entry.path for entry in contract.ENTRY_POINTS if entry.path.endswith(".py")
+        ]
+        self.assertTrue(catalogued, "no Python entry points catalogued")
+        for path in catalogued:
+            with self.subTest(entry_point=path):
+                proc = subprocess.run(
+                    [
+                        sys.executable,
+                        str(contract.plugin_path(path)),
+                        self.ABBREVIATION_PROBE,
+                    ],
+                    capture_output=True,
+                    text=True,
+                    cwd=tempfile.gettempdir(),
+                )
+                self.assertNotEqual(
+                    proc.returncode,
+                    0,
+                    f"`{path}` resolved `{self.ABBREVIATION_PROBE}` to `--help` and"
+                    " exited 0; its parser needs allow_abbrev=False",
+                )
+
+    def test_the_probe_separates_the_two_parsers(self) -> None:
+        # Guards the guard. The probe reads an exit code, and a CLI can exit 2
+        # for reasons of its own -- several here have a required mutually
+        # exclusive group that errors before any unrecognized argument is
+        # reported, so the message is not a reliable discriminator and the code
+        # is. What makes the code sufficient is that `--help` short-circuits
+        # parsing: an abbreviation that resolves exits 0 before required-argument
+        # validation ever runs. Asserted against argparse itself rather than
+        # assumed, because the whole gate rests on it.
+        strict = argparse.ArgumentParser(prog="probe", allow_abbrev=False)
+        strict.add_argument("--required-thing", required=True)
+        with self.assertRaises(SystemExit) as refused:
+            strict.parse_args([self.ABBREVIATION_PROBE])
+        self.assertEqual(refused.exception.code, 2)
+
+        lenient = argparse.ArgumentParser(prog="probe")
+        lenient.add_argument("--required-thing", required=True)
+        # The resolved `--help` prints to stdout; keep it out of the test log.
+        with contextlib.redirect_stdout(io.StringIO()):
+            with self.assertRaises(SystemExit) as resolved:
+                lenient.parse_args([self.ABBREVIATION_PROBE])
+        self.assertEqual(resolved.exception.code, 0)
+
+
 class EntryPointCatalogueIsComplete(unittest.TestCase):
     def test_every_executable_script_is_classified(self) -> None:
         # A new entry point must arrive with its mutation classification, or a