sd0xdev Smart Commit
Analyze uncommitted changes → group by cohesion → generate commit messages → output git commands (or execute directly with --execute).
Workflow
sequenceDiagram
participant C as Claude
participant U as User
C->>C: Step 1a: Detect permissions
C->>C: Step 1b: Learn commit style
C->>C: Step 1c: Identity diagnostics
C->>C: Step 1d: Signing diagnostics
C->>C: Step 1e: AI guard readiness
C->>C: Step 2: Pre-flight check (precommit)
C->>C: Step 3: Collect changes + exclude sensitive files
C->>C: Step 4: Group (high cohesion)
C->>U: Show commit plan (with Author/Signing/Guard)
U->>C: Confirm/adjust
loop Each commit group
C->>C: Read diff → generate message
C->>C: AI trailer sanitization
alt --execute mode
C->>C: Runtime validation → git commit
else manual mode
C->>U: Output git commands
end
end
C->>C: Step 6: Verification + post-commit leak detection
Git Environment Policy (applies to every step below)
Every git command this skill runs itself — read or write, either mode — resolves the same
question: which repository, tree, index, ancestry and configuration does it act on? The answer is
pinned two ways, both defined in git-environment.md § 1: the
inherited GIT_* variables are stripped, and every command carries -C "$REPO_ROOT".
Where the stripping happens depends on who runs git:
| Callee | How the policy is applied |
|---|---|
A fence running git inline |
The canonical env -u prefix, written out literally — never through a variable, because zsh does not word-split an expansion used as a command prefix |
A helper that runs its own git commands and does not strip (git-profile.sh) |
The prefix goes on the delegation, at the boundary |
A helper that strips the same list itself (smart-commit-inspect.sh, smart-commit-execute.sh) |
No prefix — the script's own unset block is the policy, and prefixing it as well would state the same rule twice in two places that could drift |
Every /bin/bash -p -- "$INSPECT" … call is the third row — all ten of them, across Steps 1b–1e,
3, 4, 5a and 6 — because smart-commit-inspect.sh unsets the list once for the whole process.
$REPO_ROOT is still re-derived in every fenced block: each block is a separate shell, so nothing
carries over.
What the skill prints is governed by § 2 of that file — the same list, the same spelling.
Step 1: Detect Permissions + Learn Style
1a. Permission Detection
Read CLAUDE.md and .claude/rules/git-workflow.md to determine mode:
| Mode | Condition | Behavior |
|---|---|---|
| manual | No --execute flag (default) |
Output commands only |
| execute | --execute flag passed |
Execute directly (with user approval via AskUserQuestion) |
Default to manual mode. Direct execution requires explicit --execute flag regardless of project git restrictions.
--execute mode: When --execute is passed, use AskUserQuestion to show the full commit plan and get explicit user approval before executing. This is a skill-level exception to git-workflow rules (same pattern as /push-ci).
1b. Learn Commit Style
REPO_ROOT=$(/usr/bin/env -u GIT_DIR -u GIT_WORK_TREE -u GIT_COMMON_DIR -u GIT_INDEX_FILE -u GIT_OBJECT_DIRECTORY -u GIT_ALTERNATE_OBJECT_DIRECTORIES -u GIT_NAMESPACE -u GIT_CEILING_DIRECTORIES -u GIT_GLOB_PATHSPECS -u GIT_ICASE_PATHSPECS -u GIT_NOGLOB_PATHSPECS -u GIT_LITERAL_PATHSPECS -u GIT_CONFIG -u GIT_CONFIG_PARAMETERS -u GIT_CONFIG_COUNT -u GIT_CONFIG_NOSYSTEM -u GIT_CONFIG_GLOBAL -u GIT_CONFIG_SYSTEM -u GIT_IMPLICIT_WORK_TREE -u GIT_GRAFT_FILE -u GIT_SHALLOW_FILE -u GIT_PREFIX -u GIT_NO_REPLACE_OBJECTS -u GIT_REPLACE_REF_BASE -u GIT_EXTERNAL_DIFF -u ALLOW_AI_COAUTHOR git rev-parse --show-toplevel && printf .) || { echo "⚠️ could not resolve the repository root — aborting" >&2; exit 1; }
REPO_ROOT=${REPO_ROOT%.}; REPO_ROOT=${REPO_ROOT%$'\n'}
INSPECT="$REPO_ROOT/.claude/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || INSPECT="$REPO_ROOT/skills/smart-commit/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || { echo "⚠️ smart-commit-inspect.sh not found — run /install-scripts" >&2; exit 1; }
/bin/bash -p -- "$INSPECT" style
Infer format, type vocabulary, subject conventions (capitalization/tense/ticket ID), and language from recent commits.
| Condition | Behavior |
|---|---|
| Commits listed | Infer the convention from them |
Non-zero exit, does not have any commits yet on stderr |
Not an error — this is the first commit of a new repository, which is a case /smart-commit exists for. Fall back to the default <type>: <subject> convention and continue. Do not apply the "stop and report the stderr" rule the later steps carry: they read state a commit depends on, this one reads history that may legitimately be empty |
| Any other non-zero exit | Stop and report the stderr — the repository could not be read |
1c. Identity Diagnostics
Shared diagnostic (preferred path):
REPO_ROOT=$(/usr/bin/env -u GIT_DIR -u GIT_WORK_TREE -u GIT_COMMON_DIR -u GIT_INDEX_FILE -u GIT_OBJECT_DIRECTORY -u GIT_ALTERNATE_OBJECT_DIRECTORIES -u GIT_NAMESPACE -u GIT_CEILING_DIRECTORIES -u GIT_GLOB_PATHSPECS -u GIT_ICASE_PATHSPECS -u GIT_NOGLOB_PATHSPECS -u GIT_LITERAL_PATHSPECS -u GIT_CONFIG -u GIT_CONFIG_PARAMETERS -u GIT_CONFIG_COUNT -u GIT_CONFIG_NOSYSTEM -u GIT_CONFIG_GLOBAL -u GIT_CONFIG_SYSTEM -u GIT_IMPLICIT_WORK_TREE -u GIT_GRAFT_FILE -u GIT_SHALLOW_FILE -u GIT_PREFIX -u GIT_NO_REPLACE_OBJECTS -u GIT_REPLACE_REF_BASE -u GIT_EXTERNAL_DIFF -u ALLOW_AI_COAUTHOR git rev-parse --show-toplevel && printf .) || { echo "⚠️ could not resolve the repository root — aborting" >&2; exit 1; }
REPO_ROOT=${REPO_ROOT%.}; REPO_ROOT=${REPO_ROOT%$'\n'}
/usr/bin/env -u GIT_DIR -u GIT_WORK_TREE -u GIT_COMMON_DIR -u GIT_INDEX_FILE -u GIT_OBJECT_DIRECTORY -u GIT_ALTERNATE_OBJECT_DIRECTORIES -u GIT_NAMESPACE -u GIT_CEILING_DIRECTORIES -u GIT_GLOB_PATHSPECS -u GIT_ICASE_PATHSPECS -u GIT_NOGLOB_PATHSPECS -u GIT_LITERAL_PATHSPECS -u GIT_CONFIG -u GIT_CONFIG_PARAMETERS -u GIT_CONFIG_COUNT -u GIT_CONFIG_NOSYSTEM -u GIT_CONFIG_GLOBAL -u GIT_CONFIG_SYSTEM -u GIT_IMPLICIT_WORK_TREE -u GIT_GRAFT_FILE -u GIT_SHALLOW_FILE -u GIT_PREFIX -u GIT_NO_REPLACE_OBJECTS -u GIT_REPLACE_REF_BASE -u GIT_EXTERNAL_DIFF -u ALLOW_AI_COAUTHOR /bin/bash -p -- "$REPO_ROOT/scripts/run-skill.sh" git-profile git-profile.sh doctor --json
The prefix goes on the delegation, not inside the helper: git-profile.sh runs its own
git commands, and stripping at the boundary is what makes it and the fallback below agree. The
interpreter is /bin/bash -p -- for the reason given once in
execute-mode.md § Why the entrypoint is spelled /bin/bash -p -- —
it applies here too, and this is the path whose JSON decides HALT versus continue.
Known limitation. In a consuming project this delegation cannot resolve git-profile.sh,
so the fallback below is the normal path there — not a broken installation. Two stacked defects,
both shared by the eleven skills that delegate through the runner; the first could be fixed here
alone, but not without diverging this one skill from the other ten, and the second would remain:
4-implementation.md § 6.
If the script succeeds, parse the JSON output:
status: "ok"→ silent continue, useeffective_identityandsigningfieldsstatus: "warn"→ display warnings fromissues[], continuestatus: "halt"→ display halt issues, stop with guidance
If the script fails (not found, parse error, non-zero exit), fall back to the diagnostic below. Infrastructure failure = warn-only; never halt on fallback path itself.
Fallback — the claim is narrow: the two still share the root derivation, git and
/bin/bash, and nothing here survives those failing. What they no longer share is the
resolution — the preferred path goes through scripts/run-skill.sh, the fallback resolves
smart-commit-inspect.sh by path, installed copy first. A missing or broken runner therefore
takes out the first and leaves the second working. Why that separation is load-bearing:
4-implementation.md § 5.
REPO_ROOT=$(/usr/bin/env -u GIT_DIR -u GIT_WORK_TREE -u GIT_COMMON_DIR -u GIT_INDEX_FILE -u GIT_OBJECT_DIRECTORY -u GIT_ALTERNATE_OBJECT_DIRECTORIES -u GIT_NAMESPACE -u GIT_CEILING_DIRECTORIES -u GIT_GLOB_PATHSPECS -u GIT_ICASE_PATHSPECS -u GIT_NOGLOB_PATHSPECS -u GIT_LITERAL_PATHSPECS -u GIT_CONFIG -u GIT_CONFIG_PARAMETERS -u GIT_CONFIG_COUNT -u GIT_CONFIG_NOSYSTEM -u GIT_CONFIG_GLOBAL -u GIT_CONFIG_SYSTEM -u GIT_IMPLICIT_WORK_TREE -u GIT_GRAFT_FILE -u GIT_SHALLOW_FILE -u GIT_PREFIX -u GIT_NO_REPLACE_OBJECTS -u GIT_REPLACE_REF_BASE -u GIT_EXTERNAL_DIFF -u ALLOW_AI_COAUTHOR git rev-parse --show-toplevel && printf .) || { echo "⚠️ could not resolve the repository root — aborting" >&2; exit 1; }
REPO_ROOT=${REPO_ROOT%.}; REPO_ROOT=${REPO_ROOT%$'\n'}
INSPECT="$REPO_ROOT/.claude/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || INSPECT="$REPO_ROOT/skills/smart-commit/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || { echo "⚠️ smart-commit-inspect.sh not found — run /install-scripts" >&2; exit 1; }
/bin/bash -p -- "$INSPECT" identity
Read the output by key and kind, never by line number. Field 1 is the key, field 2 is the
kind — value, empty or unset:
| Kind | Line shape |
|---|---|
value |
<key><TAB>value<TAB><scope><TAB><origin><TAB><value> |
empty |
<key><TAB>empty<TAB><scope><TAB><origin> — the key is set to the empty string |
unset |
<key><TAB>unset |
The kind is written by the script, never read out of the config, and every data field is
escaped (\\, \n, \t, \r, and every other C0 control byte plus DEL as \xHH — P8v) —
the value, the origin path, the four env values below, and the effective ident (P8t). So one
record is always exactly one line with a fixed field count, and no data can forge a key or a
kind. Both properties are load-bearing: without them a user.name containing a newline
emits a second line reading user.name<TAB>unset and the HALT below fires on a repository that
commits perfectly well; a GIT_AUTHOR_NAME containing one forges a whole identity record; and a
GIT_AUTHOR_NAME containing a raw ESC byte carries an ANSI/terminal escape sequence straight
into whatever prints the record later.
Records for a key arrive in git's own precedence order, so the LAST one is what git would use.
Read that line, not the presence of any line — an empty value in ~/.gitconfig overridden by a
real one in .git/config produces both an empty and a value line, and only the last of them
describes the commit that would be made.
The script then prints the four env lines, and finally the four lines that carry the
verdict — two effective, two configured. On the abort path it exits before reaching any
of them.
The env lines carry the same key/kind shape (minus scope and origin, which the environment does not have), because exported-and-empty and not-exported are different states and one blank line cannot say which:
GIT_AUTHOR_NAME<TAB>value<TAB><value> # exported, non-empty — this is what the commit uses
GIT_AUTHOR_NAME<TAB>empty # exported and empty — NOT the same as unset
GIT_AUTHOR_NAME<TAB>unset # not in the environment; config decides this role
The last four lines are the answer; everything above them says where it came from.
effective<TAB>author<TAB>value<TAB><ident> # what `git commit` would record for this role
effective<TAB>committer<TAB>unresolvable # git refuses this role outright
configured<TAB>author<TAB>yes # the ident came from something set for THIS repo
configured<TAB>committer<TAB>no # it fell through to $EMAIL or git's OS guess
effective is git var GIT_AUTHOR_IDENT / GIT_COMMITTER_IDENT, asked of git rather than rebuilt
here. git applies the whole chain — GIT_<role>_* when exported, else user.*, else EMAIL, else
a guess from the OS — then runs its ident parser and refuses in exactly the cases git commit
refuses. Two consequences a table over the config rows could not reach, both measured (§ 10.6):
| Input | The records above | effective |
|---|---|---|
user.name = " " (spaces) |
value — looks configured |
unresolvable; git commit fatals name consists only of disallowed characters |
GIT_AUTHOR_NAME=" Alice <evil@x> " |
reported verbatim | Alice evil@x <…> — git's parser strips the angle brackets, and the commit records the stripped form |
user.email unset, EMAIL exported |
user.email<TAB>unset |
the EMAIL value — a source no record here reports, because it is git's fallback, not a git config |
The trailing <unix-seconds> <tz> git appends is stripped: the question is who, not when.
configured is git -c user.useConfigOnly=true var GIT_<ROLE>_IDENT, asked the same way as
effective — not a re-derivation from the records above. git's useConfigOnly mode refuses
the whole fallback half of resolution (EMAIL, the OS guess) and succeeds only when the ident came
from something set for this repository or this invocation: user.*/author.*/committer.*
config, or an exported GIT_<role>_NAME/GIT_<role>_EMAIL. It exists because effective alone
cannot make this distinction — git's OS guess resolves successfully, so a value line looks
identical whether it came from .git/config or from uname. Measured: configured reads no for
both EMAIL and a bare OS guess alike, and that grouping is not a gap — from the operator's view
neither is a value they set for this repository, and git config --local fixes both the same way.
Oracles: P8r, P8s.
Decision logic — read the effective lines first, then use the records to locate the cause:
| Condition | Behavior |
|---|---|
Both lines are effective<TAB><role><TAB>value<TAB><ident>, each with a non-empty name and a non-empty address, and the matching configured<TAB><role> line reads yes |
Continue with no prompt. Show that ident in the commit plan; it is what git will record, whatever combination of env, config and fallback produced it |
Either line is effective<TAB><role><TAB>unresolvable |
HALT — git refuses that role, so no commit is possible. Locate the cause in the records above, and print git var GIT_AUTHOR_IDENT (or GIT_COMMITTER_IDENT) as the command that shows git's own message |
An effective ident whose address is empty — Name <> |
HALT — nothing fails here; git commits and the commit simply carries no attribution, which is the outcome CLAUDE.md rule 3 exists to prevent. Measured P8m: an empty GIT_AUTHOR_EMAIL/GIT_COMMITTER_EMAIL or an empty user.email both land here |
Provenance: an effective line resolved to a value (not already caught by the empty-address row above), and the matching configured<TAB><role> line reads no |
HALT — the ident did not come from anything set for this repository: it is $EMAIL or git's OS guess, and configured cannot tell those two apart (§ 10.8) — nor does the fix need it to: git config --local user.name "..." / user.email "..." (or the author.*/committer.* equivalent) covers both. This is a policy halt, not a git refusal: git commits happily on either source, user.useConfigOnly=true is what draws the line |
Two or more value records for the same config key with different values |
AskUserQuestion: list candidate profiles, user selects once. effective already says which one git picked — the prompt confirms that is the intended one |
Conflict (the row above) + a CI<TAB>value<TAB>... record |
HALT (fail-closed) — output fix guidance, do not silently inherit. identity emits this record the same way it emits the GIT_AUTHOR_*/GIT_COMMITTER_* ones, so the condition is readable from the fallback path's own output rather than assumed (P8t2) |
identity: could not read … on stderr, non-zero exit |
Stop and report the stderr. The config was unreadable; this is not an unset key and the setup guidance above would be the wrong instruction. Any stdout printed before the abort is not an answer |
Locating the cause. None of these rows is a verdict on its own — the verdict is above. They say
which input produced it, and therefore which fix can reach it. An env line outranks config for the
role it names: git takes GIT_AUTHOR_NAME when exported and falls back to user.name only when
it is not, so a role resolved from the environment is never fixed by git config.
The same outranking happens one level down, inside config itself: author.*/committer.* outrank
user.* for their own role when both are set (git >= 2.31) — author.name wins for the author
role even if user.name is also configured, and symmetrically for committer.*. Both keys can show
an equally-valid value record below; the records alone do not say which file git actually used.
effective does — it is the row that says which config key's file to edit, not the mere presence of
a value record for user.*.
| Record | What it means for the role |
|---|---|
GIT_AUTHOR_NAME<TAB>value<TAB><value> |
Exported and non-empty — this is the input; git config cannot reach it, unset can |
GIT_AUTHOR_NAME<TAB>empty |
Exported and empty. For a NAME this is why effective says unresolvable; for an EMAIL it is why the address is <>. Fixed by unset, never by git config |
GIT_AUTHOR_NAME<TAB>unset |
Not in the environment — config decides this role, so read its key below |
<key><TAB>value<TAB><scope><TAB><origin><TAB><value> |
The configured input, and <origin> is the file to edit. Only the last record for the key is the one git used |
<key><TAB>empty<TAB><scope><TAB><origin> |
Set to the empty string in <origin>. As the last record for the key this is a cause; earlier in the list it is shadowed and irrelevant |
<key><TAB>unset |
Not configured anywhere. If no env var covers the role either, effective is one of two measured outcomes: it resolved from EMAIL or git's OS guess (the provenance HALT above), or it reads unresolvable — configuring the OTHER role's author.*/committer.* alone is enough to disable the OS guess for THIS one, even though this role itself has no record above saying so (P8s, § 10.8) |
Oracles: P8, P8b–P8u; the measured forgeries and misreads:
4-implementation.md § 10.
Design principles:
- Diagnostic, not override: Do not use
git -c user.name=...to override. RespectincludeIfsettings. - Ask git, do not model git: the verdict comes from
git var; the records exist to locate a cause, never to re-derive the answer. - Interrupt only on anomaly: Normal identity resolution produces no prompt.
- Conflict ≠ multiple sources:
includeIfproducing multiple config sources that resolve to the same value = normal.
1d. Signing Diagnostics
REPO_ROOT=$(/usr/bin/env -u GIT_DIR -u GIT_WORK_TREE -u GIT_COMMON_DIR -u GIT_INDEX_FILE -u GIT_OBJECT_DIRECTORY -u GIT_ALTERNATE_OBJECT_DIRECTORIES -u GIT_NAMESPACE -u GIT_CEILING_DIRECTORIES -u GIT_GLOB_PATHSPECS -u GIT_ICASE_PATHSPECS -u GIT_NOGLOB_PATHSPECS -u GIT_LITERAL_PATHSPECS -u GIT_CONFIG -u GIT_CONFIG_PARAMETERS -u GIT_CONFIG_COUNT -u GIT_CONFIG_NOSYSTEM -u GIT_CONFIG_GLOBAL -u GIT_CONFIG_SYSTEM -u GIT_IMPLICIT_WORK_TREE -u GIT_GRAFT_FILE -u GIT_SHALLOW_FILE -u GIT_PREFIX -u GIT_NO_REPLACE_OBJECTS -u GIT_REPLACE_REF_BASE -u GIT_EXTERNAL_DIFF -u ALLOW_AI_COAUTHOR git rev-parse --show-toplevel && printf .) || { echo "⚠️ could not resolve the repository root — aborting" >&2; exit 1; }
REPO_ROOT=${REPO_ROOT%.}; REPO_ROOT=${REPO_ROOT%$'\n'}
INSPECT="$REPO_ROOT/.claude/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || INSPECT="$REPO_ROOT/skills/smart-commit/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || { echo "⚠️ smart-commit-inspect.sh not found — run /install-scripts" >&2; exit 1; }
/bin/bash -p -- "$INSPECT" signing
Decision logic:
signing answers in the same shape as Step 1c — <key><TAB><kind>[…] for
commit.gpgsign, user.signingkey and gpg.format, in that order, read by key and never by
line number, with the last record for a key winning. It used to print three bare lines, which a
multi-line value silently shifted (§ 10.2). Oracles: P9, P9b, P9c.
commit.gpgsign is read as a boolean, so its value field is only ever the literal true or
false: git normalises 1/yes/on/True, and a valueless gpgsign — which git reads as
true — is no longer byte-identical to gpgsign =, which git reads as false. That
distinction cannot be recovered from the raw string, which is why it is git that converts and not
a row in this table. commit.gpgsign<TAB>empty is therefore unreachable, and a non-boolean value
is a read failure rather than a verdict. Oracles: P9d–P9f; measurements: § 10.3.
| Condition (per key, last record) | Behavior |
|---|---|
commit.gpgsign<TAB>value…true + user.signingkey<TAB>value |
Display Signing: enabled (<gpg.format>) |
commit.gpgsign<TAB>value…true + user.signingkey<TAB>unset or user.signingkey<TAB>empty |
⚠️ Warning: signing enabled but key not configured |
commit.gpgsign<TAB>value…false |
Display Signing: disabled (commit.gpgsign=false). Distinct from the row below: this is an explicit off, so --sign contradicts a stated intent rather than an absent one |
commit.gpgsign<TAB>unset |
Display Signing: not configured (inherit) |
commit.gpgsign<TAB>empty |
Unreachable under --type=bool. Treat as a read failure, not as false — reaching it means the script was invoked without the flag, so the verdict would be untrustworthy either way |
gpg.format<TAB>value |
Use it as the <gpg.format> shown above |
gpg.format<TAB>unset or gpg.format<TAB>empty |
Use git's default, gpg. The script reports the fact; applying the default is this step's job |
signing: could not read … on stderr, non-zero exit |
Stop and report the stderr — same reasoning as Step 1c. This is also where a non-boolean commit.gpgsign lands, and it is not stricter than git: git's own --bool read exits 128 on the same config, so a git commit would fail too |
--execute mode signing failure |
Immediate stop + fix guidance |
Post-commit visibility (--execute mode):
REPO_ROOT=$(/usr/bin/env -u GIT_DIR -u GIT_WORK_TREE -u GIT_COMMON_DIR -u GIT_INDEX_FILE -u GIT_OBJECT_DIRECTORY -u GIT_ALTERNATE_OBJECT_DIRECTORIES -u GIT_NAMESPACE -u GIT_CEILING_DIRECTORIES -u GIT_GLOB_PATHSPECS -u GIT_ICASE_PATHSPECS -u GIT_NOGLOB_PATHSPECS -u GIT_LITERAL_PATHSPECS -u GIT_CONFIG -u GIT_CONFIG_PARAMETERS -u GIT_CONFIG_COUNT -u GIT_CONFIG_NOSYSTEM -u GIT_CONFIG_GLOBAL -u GIT_CONFIG_SYSTEM -u GIT_IMPLICIT_WORK_TREE -u GIT_GRAFT_FILE -u GIT_SHALLOW_FILE -u GIT_PREFIX -u GIT_NO_REPLACE_OBJECTS -u GIT_REPLACE_REF_BASE -u GIT_EXTERNAL_DIFF -u ALLOW_AI_COAUTHOR git rev-parse --show-toplevel && printf .) || { echo "⚠️ could not resolve the repository root — aborting" >&2; exit 1; }
REPO_ROOT=${REPO_ROOT%.}; REPO_ROOT=${REPO_ROOT%$'\n'}
INSPECT="$REPO_ROOT/.claude/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || INSPECT="$REPO_ROOT/skills/smart-commit/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || { echo "⚠️ smart-commit-inspect.sh not found — run /install-scripts" >&2; exit 1; }
/bin/bash -p -- "$INSPECT" signature
It prints one character, git's %G? alphabet (git log manual, git 2.54.0), and nothing else.
Read the character, not the line count — and treat an empty stdout with a non-zero exit as its own
state, because a repository with no commit to inspect is not an unsigned commit:
%G? |
git's meaning | Behavior |
|---|---|---|
G |
good (valid) signature | Display Signed: good |
U |
good signature, unknown validity | Display Signed: good (signer key not trusted locally) — a local keyring fact, not a defect in the commit |
X |
good signature that has expired | ⚠️ Warning: verifies locally, and this is the state a server-side signature policy rejects |
Y |
good signature made by an expired key | ⚠️ Warning — same reason as X |
R |
good signature made by a revoked key | ⚠️ Warning — same reason as X, and the most likely of the three to be rejected |
E |
signature cannot be checked (e.g. missing key) | ⚠️ Warning: says the local keyring cannot verify it, not that the commit is unsigned |
B |
bad signature | Stop and report. The commit carries a signature that does not verify |
N |
no signature | Benign when Step 1d reported signing off or unset. Stop when this batch was meant to be signed (commit.gpgsign=true, no --no-sign): the commit just made is unsigned |
| (empty stdout, non-zero exit) | — | Not a verdict. The repository has no commit to inspect (measured: unborn repo → rc 128, empty stdout) or could not be read. Report the stderr; do not read it as N |
Oracles: P11, P11b.
Signing override flags (--sign / --no-sign):
| Flag | Effect | Git flag |
|---|---|---|
--sign |
Force signing for this batch | -S on each git commit |
--no-sign |
Disable signing for this batch | --no-gpg-sign on each git commit |
| Both | Error — mutually exclusive | Halt with error message |
| Neither | Inherit from commit.gpgsign config |
(default behavior) |
When --sign or --no-sign is used, AskUserQuestion to confirm the override and warn about potential branch protection / CI policy conflicts.
1e. AI Guard Readiness
# Hook resolution is git's answer, not ours: `--git-path` honours core.hooksPath
# (including `~` and `%(prefix)`) and resolves linked worktrees. Why the three states
# differ: test/scripts/smart-commit-inspect.test.js `P6`. Why a resolution FAILURE is
# a fourth state and not `guard:missing`: smart-commit-inspect.sh, `guard`.
REPO_ROOT=$(/usr/bin/env -u GIT_DIR -u GIT_WORK_TREE -u GIT_COMMON_DIR -u GIT_INDEX_FILE -u GIT_OBJECT_DIRECTORY -u GIT_ALTERNATE_OBJECT_DIRECTORIES -u GIT_NAMESPACE -u GIT_CEILING_DIRECTORIES -u GIT_GLOB_PATHSPECS -u GIT_ICASE_PATHSPECS -u GIT_NOGLOB_PATHSPECS -u GIT_LITERAL_PATHSPECS -u GIT_CONFIG -u GIT_CONFIG_PARAMETERS -u GIT_CONFIG_COUNT -u GIT_CONFIG_NOSYSTEM -u GIT_CONFIG_GLOBAL -u GIT_CONFIG_SYSTEM -u GIT_IMPLICIT_WORK_TREE -u GIT_GRAFT_FILE -u GIT_SHALLOW_FILE -u GIT_PREFIX -u GIT_NO_REPLACE_OBJECTS -u GIT_REPLACE_REF_BASE -u GIT_EXTERNAL_DIFF -u ALLOW_AI_COAUTHOR git rev-parse --show-toplevel && printf .) || { echo "⚠️ could not resolve the repository root — aborting" >&2; exit 1; }
REPO_ROOT=${REPO_ROOT%.}; REPO_ROOT=${REPO_ROOT%$'\n'}
INSPECT="$REPO_ROOT/.claude/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || INSPECT="$REPO_ROOT/skills/smart-commit/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || { echo "⚠️ smart-commit-inspect.sh not found — run /install-scripts" >&2; exit 1; }
/bin/bash -p -- "$INSPECT" guard
Decision logic:
| Condition | Behavior |
|---|---|
| Hook installed + executable | Display AI guard: active |
| Hook not installed | ⚠️ Warning (non-blocking): suggest install (/install-scripts commit-msg-guard then cp .claude/scripts/commit-msg-guard.sh <hooks-path>/commit-msg && chmod +x <hooks-path>/commit-msg) |
| Hook exists but not executable | ⚠️ Warning: suggest chmod +x <hook-path> |
Non-zero exit (guard: could not resolve the hooks path — aborting) |
Stop and report the stderr. This is not guard:missing: that verdict says "install the hook", which is the wrong instruction when the truth is that the repository could not be read |
Important: Hook installation is NOT a blocker for --execute mode — runtime validation (Step 5c) is an independent safety layer.
Step 2: Pre-flight Check
Check precommit status by change type — structural .md under skills/ has test coverage (skills-schema.test.js) that catches reference errors CI would find.
| Change Type | Required Check | Rationale |
|---|---|---|
Code files (.ts/.js/.py/.go/.rs etc.) |
/precommit or /precommit-fast passed |
Code correctness + lint |
Structural .md (skills/**) |
/precommit-fast passed |
Schema/ref tests cover SKILL.md structure |
Other .md (README, docs/) |
/codex-review-doc passed (per CLAUDE.md) |
No structural tests; doc review sufficient |
| Trivial whitespace in non-code files | Skip allowed | No test coverage expected. Comment edits inside code files are not here — CLAUDE.md classifies them conservatively as code (comments can carry compiler/lint/build directives), so they follow the code-file row above |
Default: advisory, not a gate. /smart-commit can be invoked before review has ever run — the
terminal completion invariant (rules/auto-loop.md) constrains the moment a change is declared
complete, not the moment this skill is called — so on the common path, where auto-loop already ran
review and precommit and noted the verdict, re-halting on the same evidence is friction, not
added safety.
| Status | Default action | --strict-preflight action |
|---|---|---|
| Required check passed in current session after last edit | Continue (silent) | Continue (silent) |
| Not run, stale, or uncertain | ⚠️ Warn and continue — name the missing/stale gate(s) and the exact command to close each one | Halt — ask user to run the required check first |
--strict-preflight: opt-in flag restoring the original Halt behavior for the not-fresh row —
the passed row already continues silently in both columns, so there is nothing for the flag to
change there. Manual and --execute modes honor it identically — the flag changes Step 2's own
decision, not which mode invoked the skill.
The warning must be actionable — name the gate and the command, not just "not passed":
⚠️ Pre-flight: 2 of 3 changed files have no fresh check this session.
- src/service/rpc.ts (code): run /precommit-fast or /precommit
- docs/features/x/2-tech-spec.md (other .md): run /codex-review-doc
Continuing without --strict-preflight. Pass --strict-preflight to halt on this instead.
Freshness: A "passed" result is only valid if it ran after the most recent file edits in this session. Stale results from earlier in the session do not count — freshness decides which row above applies, not whether that row warns or halts.
Policy note: see "Default: advisory, not a gate" above for the rationale — this only relaxes
this skill's own extra pass over verdicts already noted; @rules/auto-loop.md's gates are
unaffected, and /precommit//codex-review-doc are not exempted for anyone still passing through
them. Worth stating plainly: the Stop hook is a reminder, never a block
(hook-lightweighting § 3.2 — it emits markdown and exits 0), so Step 2's own Halt is the only
blocking barrier this side of the git hooks between an unreviewed edit and a commit.
--strict-preflight is how to raise that barrier, not a stylistic preference.
Fast vs full: /precommit-fast runs test:fast; CI runs test:ci. Deletions (skills, scripts) can leave orphaned test files that only fail in CI, so when only /precommit-fast ran, output: ⚠️ Only fast tests ran. If you deleted files, consider /precommit (full suite) to catch orphaned imports before commit.
Step 3: Collect Changes
# Root-relative by construction — `git status --short` reports paths relative to the
# CURRENT directory, so a subdirectory would emit pathspecs a later `-C` misresolves.
REPO_ROOT=$(/usr/bin/env -u GIT_DIR -u GIT_WORK_TREE -u GIT_COMMON_DIR -u GIT_INDEX_FILE -u GIT_OBJECT_DIRECTORY -u GIT_ALTERNATE_OBJECT_DIRECTORIES -u GIT_NAMESPACE -u GIT_CEILING_DIRECTORIES -u GIT_GLOB_PATHSPECS -u GIT_ICASE_PATHSPECS -u GIT_NOGLOB_PATHSPECS -u GIT_LITERAL_PATHSPECS -u GIT_CONFIG -u GIT_CONFIG_PARAMETERS -u GIT_CONFIG_COUNT -u GIT_CONFIG_NOSYSTEM -u GIT_CONFIG_GLOBAL -u GIT_CONFIG_SYSTEM -u GIT_IMPLICIT_WORK_TREE -u GIT_GRAFT_FILE -u GIT_SHALLOW_FILE -u GIT_PREFIX -u GIT_NO_REPLACE_OBJECTS -u GIT_REPLACE_REF_BASE -u GIT_EXTERNAL_DIFF -u ALLOW_AI_COAUTHOR git rev-parse --show-toplevel && printf .) || { echo "⚠️ could not resolve the repository root — aborting" >&2; exit 1; }
REPO_ROOT=${REPO_ROOT%.}; REPO_ROOT=${REPO_ROOT%$'\n'}
INSPECT="$REPO_ROOT/.claude/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || INSPECT="$REPO_ROOT/skills/smart-commit/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || { echo "⚠️ smart-commit-inspect.sh not found — run /install-scripts" >&2; exit 1; }
/bin/bash -p -- "$INSPECT" collect
Classify changes:
Check unmerged first — it pre-empts every row below it, including staged. UU/AA/DD
(and the AU/UD/UA/DU variants) all carry a non-blank column 1, so reading the table
top-to-bottom without this note routes an unresolved conflict into "staged" — group it with no
git add needed, then git commit a path the merge never finished. Measured: UU after a real
git merge conflict.
| Type | Status code | Description |
|---|---|---|
| unmerged | UU, AA, DD, AU, UD, UA, DU |
Stop, do not group. An unresolved merge conflict — surface it to the user; it needs git add only after the conflict markers are resolved by hand |
| staged | any other code in column 1 | Already git add-ed |
| modified | M |
Tracked but unstaged |
| typechange | T in either column (e.g. T, T) |
File replaced by a different type at the same path — most commonly a regular file swapped for a symlink. Group like modified; the path is unambiguous, there is nothing to resolve |
| untracked | ?? |
New files (decide whether to include) |
| deleted | D in either column |
Deleted files |
| renamed | R |
Two paths on one line: R <old> -> <new>. Group by <new> — that is the path a later git add -- ':(literal)<path>' must name |
| copied | C |
Same two-path shape. Reachable whenever status.renames=copies is inherited from any config scope, so it is not a case this skill may assume away |
The -> in those two rows is a separator, not part of a path, and telling the two apart is
not a guess: git quotes any path containing ->, in every status shape. Measured on an untracked
file, a modified file, and a rename in each direction — the four lines collect produced:
?? "a -> b.txt"
M "a -> b.txt"
R src.txt -> "dst -> x.txt"
R "a -> b.txt" -> plain.txt
So an unquoted -> is always the rename separator. Splitting on it without that check is
what turns a -> b.txt into two pathspecs that match nothing.
Exclusion rules (warn user, do not include):
.env* | *.pem | *.key | *.p12 | id_rsa* | .aws/credentials | *.secret
credentials.json | .npmrc | token.txt | node_modules/ | dist/ | .cache/ | .gitignore'd
Partial-staged detection: If a file has both staged and unstaged changes (MM in git status), warn user and ask them to resolve first.
--scope filtering: When --scope <path> is specified, only include changes under that path. Apply after collecting all changes:
REPO_ROOT=$(/usr/bin/env -u GIT_DIR -u GIT_WORK_TREE -u GIT_COMMON_DIR -u GIT_INDEX_FILE -u GIT_OBJECT_DIRECTORY -u GIT_ALTERNATE_OBJECT_DIRECTORIES -u GIT_NAMESPACE -u GIT_CEILING_DIRECTORIES -u GIT_GLOB_PATHSPECS -u GIT_ICASE_PATHSPECS -u GIT_NOGLOB_PATHSPECS -u GIT_LITERAL_PATHSPECS -u GIT_CONFIG -u GIT_CONFIG_PARAMETERS -u GIT_CONFIG_COUNT -u GIT_CONFIG_NOSYSTEM -u GIT_CONFIG_GLOBAL -u GIT_CONFIG_SYSTEM -u GIT_IMPLICIT_WORK_TREE -u GIT_GRAFT_FILE -u GIT_SHALLOW_FILE -u GIT_PREFIX -u GIT_NO_REPLACE_OBJECTS -u GIT_REPLACE_REF_BASE -u GIT_EXTERNAL_DIFF -u ALLOW_AI_COAUTHOR git rev-parse --show-toplevel && printf .) || { echo "⚠️ could not resolve the repository root — aborting" >&2; exit 1; }
REPO_ROOT=${REPO_ROOT%.}; REPO_ROOT=${REPO_ROOT%$'\n'}
INSPECT="$REPO_ROOT/.claude/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || INSPECT="$REPO_ROOT/skills/smart-commit/scripts/smart-commit-inspect.sh"
[ -r "$INSPECT" ] || { echo "⚠️ smart-commit-inspect.sh not found — run /install-scripts" >&2; exit 1; }
/bin/bash -p -- "$INSPECT" scope '<path>'
Substitute the --scope argument for '<path>', single-quoted, and a path containing an apostrophe is written ''' at that position (the same rule references/git-environment.md § 2 gives for printed commands; without it the fence is syntactically broken rather than merely wrong) — written out rather than passed
as "$SCOPE_PATH" because each fence is one tool call, so one shell: a variable this fence
never assigned expands to the empty string, the script refuses it (scope: the path must not be empty, exit 2) and the scoping the caller asked for never happens — loudly, but it still does not
happen. Same rule Step 5c.3 states for "$EXECUTE".
Exclude changes outside the scope path. --scope takes a path, never a glob: every operand
carries :(literal), so *.ts matches a file literally named *.ts and nothing else. If no
changes remain after filtering → report "No changes under <path>" and stop.
If no changes → report "No uncommitted changes" and stop.
A non-zero exit from collect is not an empty change set. The subcommand accumulates the
status of all three git commands, so one failing among two that printed still surfaces — output
with no status lines can mean a broken git status, not a clean tree. Report the stderr and stop;
never take the "no changes" branch on a non-zero exit. A successful empty answer is a real clean
tree: collect, status and scope all pin --untracked-files=all, so showUntrackedFiles=no
cannot fake one, and a wholly-untracked directory is enumerated per file rather than collapsed to
?? dir/ — which is what lets the exclusion list below match a secret by filename. Two kinds
of directory stay collapsed because git will not traverse them, an embedded repository and a
symlink to a directory; both stage as a gitlink or a symlink and carry no file content, so nothing
hides inside them. This holds wherever those three subcommands are used, Step 6's re-check
included. The cost is real and accepted: a large untracked tree that is not gitignored now
lists one line per file. Triggers, measurements and the trade-off: P15, P17, P17b, P17c in
test/scripts/smart-commit-inspect.test.js and
4-implementation.md § 9.
Selection is live git status, whole tree. Every uncommitted change collect reported — minus
the exclusion rules and any --scope filter above — is a candidate for grouping. There is no
session-scoped filter: the hook-maintained session_commit_scope store retired with the state
machine that wrote it (hook-lightweighting § 3.4), and nothing now records which files this
session touched. What replaces it is not a narrower selection but a visible one: the Commit Plan
in Step 4 lists every file it is about to group, and the user confirming that plan is the
selection decision. A file the user does not want committed is removed at the plan, or fenced out
up front with --scope.
Step 4: Group (High Cohesion)
Each group should form
…(truncated)