git:20260814.b9c6b5a to git:20260821.1a05a80

17 added, 17 removed. Audit A to A.

---

name: test-impact

description: "Identify which tests need rerunning after a code change — traces static call graph (function-level) or import graph (module-level) to find affected test files, then emits a ready-to-run pytest command. TRIGGER when: user asks which tests are affected by a change; phrases: \"which tests are affected\", \"what tests cover this\", \"test impact of\", \"what tests to rerun\"."

argument-hint: "<module::symbol | module> [--no-mocks]"

allowed-tools: Bash, Write, AskUserQuestion

model: haiku

effort: low

---

- Identifies minimal test set affected by changing a function or module. Uses codemap static analysis — no test execution needed. Result: ready-to-run `pytest` command covering only impacted tests.

+ Find minimal tests affected by function/module change via Codemap static analysis; run no tests. Return ready `pytest` command for impacted tests only.

Two input modes:

- - **Function-level** (`module::symbol`) — BFS over reverse call graph; finds every test calling changed function, directly or transitively. Includes tests mocking the symbol (`mock_patches`).

- - **Module-level** (bare `module`) — BFS over reverse import graph; finds every test importing module through any chain. Includes tests mocking any symbol in module.

+ - **Function-level** (`module::symbol`) — reverse-call-graph BFS; all direct/transitive test callers + symbol mocks (`mock_patches`).

+ - **Module-level** (bare `module`) — reverse-import-graph BFS; all tests importing through any chain + any module-symbol mocks.

- `not_covered`: dynamic dispatch, hook callbacks, string-dispatch callers — same blind spot as `fn-blast`. Surface caveat, log gap.

+ `not_covered`: dynamic dispatch, hook callbacks, string-dispatch callers; same blind spot as `fn-blast`. Surface caveat; log gap.

NOT for: finding all callers of a function (use `/codemap-py:query-code fn-rdeps <module::symbol> --exclude-tests`); querying module deps or blast radius (use `/codemap-py:query-code`); running/executing tests (identified here, not executed).

</objective>

- **$ARGUMENTS**: `<qname> [--no-mocks]`

- `qname` — `module::symbol` (function-level) or bare dotted module (module-level)

- `--no-mocks` — exclude mock-only test files (no call/import path)

- Omitted → AskUserQuestion in Step 1

</inputs>

## Step 0 — Ensure index

```bash

# timeout: 10000

export CSID="${CLAUDE_CODE_SESSION_ID:-$PPID}"

_CM_PROJ=$(git rev-parse --show-toplevel 2>/dev/null | xargs basename 2>/dev/null || basename "$PWD")

_IDX="${CODEMAP_INDEX_DIR:-.cache/codemap}"

INDEX="${_IDX}/${_CM_PROJ}.json"

# dispatcher, not the scan-query/scan-index aliases — aliases lease in-engine too, but skip the dispatcher's interpreter probe (exit 127) and are deprecated shims, removed no earlier than 1.0.0

# PATH-literal invocation everywhere below — expansion-bearing form matches no bare-name allow prefix; the plugin's absolute bin/codemap-py stays the interactive fallback

command -v codemap-py >/dev/null 2>&1 || { echo "codemap-py not on PATH — install the codemap-py plugin, or invoke its bin/codemap-py launcher as one standalone command"; exit 1; }

[ ! -f "$INDEX" ] && echo "No index found — will build via codemap-py index"

```

- Auto-build opt-out via `SCAN_NO_AUTOBUILD=1` (index used exactly as-is — no refresh, no full build); build wall-time echoed when it runs, keeps build cost separable from query cost.

+ `SCAN_NO_AUTOBUILD=1` opts out: use index unchanged; no refresh/full build. Echo build wall-time when run, separating build/query cost.

- If `$INDEX` not found:

- - `SCAN_NO_AUTOBUILD=1` set → print `! codemap index missing and SCAN_NO_AUTOBUILD=1 — refusing to auto-build. Build it manually first: /codemap-py:scan-codebase` and exit 1.

- - otherwise → run `codemap-py index` in the foreground (wait until it finishes) then continue. (Not the `codemap-py:scan-codebase` skill — it is `disable-model-invocation:true`, user-slash-only; build via the gated `codemap-py index` dispatcher.)

+ If `$INDEX` missing:

+ - With `SCAN_NO_AUTOBUILD=1`: print `! codemap index missing and SCAN_NO_AUTOBUILD=1 — refusing to auto-build. Build it manually first: /codemap-py:scan-codebase`; exit 1.

+ - Otherwise run foreground `codemap-py index`; wait, then continue. Never invoke `codemap-py:scan-codebase`: `disable-model-invocation:true`, user-slash-only. Build through gated `codemap-py index` dispatcher.

If index already exists:

```bash

# timeout: 30000

_CM_PROJ=$(git rev-parse --show-toplevel 2>/dev/null | xargs basename 2>/dev/null || basename "$PWD")

_IDX="${CODEMAP_INDEX_DIR:-.cache/codemap}"

if [ "${SCAN_NO_AUTOBUILD:-0}" = "1" ]; then

echo "[codemap] SCAN_NO_AUTOBUILD=1 — using existing index as-is (no refresh)"

else

_CM_BUILD_T0=$(date +%s)

# export, not an inline env prefix — a prefix puts an expansion ahead of the binary, so the command no longer matches a bare-name allow prefix

export CODEMAP_INDEX_DIR="${_IDX}" # forward to the build; ensures it writes to same path as INDEX

codemap-py index --incremental \

&& echo "[codemap] index built in $(( $(date +%s) - _CM_BUILD_T0 ))s" \

|| printf "⚠ codemap-py index --incremental failed — index may be stale; continuing\n"

```

- After the build or incremental refresh, re-verify index still present:

+ After build/refresh, re-verify index:

```bash

# timeout: 5000

_CM_PROJ=$(git rev-parse --show-toplevel 2>/dev/null | xargs basename 2>/dev/null || basename "$PWD")

_IDX="${CODEMAP_INDEX_DIR:-.cache/codemap}"

INDEX="${_IDX}/${_CM_PROJ}.json"

[ -f "$INDEX" ] || { printf "! Index not found after refresh at %s — check CODEMAP_INDEX_DIR or re-run /codemap-py:scan-codebase\n" "$INDEX"; exit 1; }

```

## Step 1 — Parse arguments

- Extract `QNAME` and `NO_MOCKS` flag from `$ARGUMENTS`.

+ Extract `QNAME` + `NO_MOCKS` from `$ARGUMENTS`.

```bash

# timeout: 5000

export CSID="${CLAUDE_CODE_SESSION_ID:-$PPID}"

_CM_PROJ=$(git rev-parse --show-toplevel 2>/dev/null | xargs basename 2>/dev/null || basename "$PWD")

ARGS="${ARGUMENTS:-}"

QNAME=$(echo "$ARGS" | awk '{print $1}')

MOCKS_FLAG=$(echo "$ARGS" | grep -q -- "--no-mocks" && echo "--no-mocks" || echo "")

echo "$QNAME" > "${TMPDIR:-/tmp}/codemap-${_CM_PROJ}-ti-qname-${CSID}"

echo "$MOCKS_FLAG" > "${TMPDIR:-/tmp}/codemap-${_CM_PROJ}-ti-mocks-${CSID}"

```

- If `$ARGUMENTS` empty → `AskUserQuestion`: "Which function or module changed?" Options: (a) Enter `module::symbol` for function-level · (b) Enter bare module name for module-level · (c) Cancel — exit without running test-impact analysis. After the user answers, set `QNAME` from the answer and write it to the tmpfile before proceeding to Step 2:

+ If `$ARGUMENTS` empty, use `AskUserQuestion`: "Which function or module changed?" Options: (a) Enter `module::symbol` for function-level; (b) Enter bare module for module-level; (c) Cancel, exit without analysis. After answer, set `QNAME`; write tmpfile before Step 2:

```bash

# timeout: 5000

export CSID="${CLAUDE_CODE_SESSION_ID:-$PPID}"

_CM_PROJ=$(git rev-parse --show-toplevel 2>/dev/null | xargs basename 2>/dev/null || basename "$PWD")

QNAME="<answer from AskUserQuestion>"

echo "$QNAME" > "${TMPDIR:-/tmp}/codemap-${_CM_PROJ}-ti-qname-${CSID}"

```

- **Multi-symbol guard**: `$ARGUMENTS` may contain multiple space-separated tokens (e.g. `mypackage.auth::validate mypackage.auth::parse`). `awk '{print $1}'` silently truncates to first. If `$ARGUMENTS` has more than one token after stripping `--no-mocks`, print `⚠ test-impact accepts one symbol at a time — using first token only: $QNAME. Run separately for each remaining symbol.`

+ **Multi-symbol guard**: `$ARGUMENTS` may contain multiple tokens (for example `mypackage.auth::validate mypackage.auth::parse`); `awk '{print $1}'` silently keeps first. If more than one after removing `--no-mocks`, print `⚠ test-impact accepts one symbol at a time — using first token only: $QNAME. Run separately for each remaining symbol.`

## Step 2 — Run test-impact query

```bash

# timeout: 10000

export CSID="${CLAUDE_CODE_SESSION_ID:-$PPID}"

_CM_PROJ=$(git rev-parse --show-toplevel 2>/dev/null | xargs basename 2>/dev/null || basename "$PWD")

IFS= read -r QNAME < "${TMPDIR:-/tmp}/codemap-${_CM_PROJ}-ti-qname-${CSID}" 2>/dev/null || QNAME=""

IFS= read -r MOCKS_FLAG < "${TMPDIR:-/tmp}/codemap-${_CM_PROJ}-ti-mocks-${CSID}" 2>/dev/null || MOCKS_FLAG=""

_TI_ERR="${TMPDIR:-/tmp}/codemap-${_CM_PROJ}-ti-stderr-${CSID}"

# capture stderr to a file, never 2>/dev/null — a swallowed diagnostic is what rendered a broken index as "no affected tests"

RESULT=$(codemap-py query test-impact "$QNAME" $MOCKS_FLAG 2>"$_TI_ERR")

_TI_RC=$?

if [ "$_TI_RC" -ne 0 ]; then

printf "! test-impact query failed (exit %d) — this is NOT a 'no affected tests' result; do not report an empty test set\n" "$_TI_RC" >&2

[ -s "$_TI_ERR" ] && sed -n '1,5p' "$_TI_ERR" >&2

printf "Rebuild with /codemap-py:scan-codebase, then re-run.\n" >&2

exit 1

# one parse, no per-field `|| echo` defaults — a default here forges total=0 out of an unparsable payload

printf '%s' "$RESULT" | python3 -c "

import json, sys

base, suf = sys.argv[1], sys.argv[2]

try:

payload = json.loads(sys.stdin.read())

except ValueError:

sys.stderr.write('! test-impact returned non-JSON output — NOT an empty result; rebuild via /codemap-py:scan-codebase and re-run\n')

raise SystemExit(1)

index = payload.get('index') or {}

fields = {

'not-covered': json.dumps(index.get('not_covered') or []),

'hint': index.get('hint') or '',

'total': str(len(payload.get('test_files') or [])),

'pytest-cmd': payload.get('pytest_cmd') or '',

}

for name, value in fields.items():

with open(base + name + suf, 'w') as handle:

handle.write(value + '\n')

" "${TMPDIR:-/tmp}/codemap-${_CM_PROJ}-ti-" "-${CSID}" || exit 1

```

- Parse JSON output from `$RESULT`:

+ Parse `$RESULT` JSON:

- `test_files` — list of test file paths

- `pytest_cmd` — ready-to-run command

- `via_call` / `via_mock` — breakdown of how tests were found

- `index.not_covered` — surface as caveat if non-empty

- `index.hint` — include as suggestion

- **Loud-failure contract**: query failure and empty result are different outcomes and must never collapse into one another. The block above enforces both halves in shell, not prose — a non-zero exit code stops the skill (exit 1) before any field is read, and an unparsable payload exits 1 from the parser. Neither path may be reported as "no affected tests". A false empty test set on a broken index is the worst output this skill can emit.

+ **Loud-failure contract**: query failure ≠ empty result. Shell enforces it: nonzero exits skill `1` before field reads; unparsable payload exits parser `1`. Neither may become "no affected tests". False empty set from broken index is worst possible output.

- **haiku JSON parse guard**: `codemap-py query` JSON output may be prefixed/suffixed with log/warning lines under the haiku model. Always extract JSON by piping stdin into `python3 -c "import json, sys; ..."` — never assume raw output is valid JSON, and never add a per-field `|| echo "0"` / `|| echo "[]"` fallback, which manufactures a benign-looking default out of a failed parse.

+ **haiku JSON guard**: output may include log/warning prefix/suffix. Always pipe stdin into `python3 -c "import json, sys; ..."`; never assume raw valid JSON. Never add per-field `|| echo "0"` / `|| echo "[]"`: failed parse must not manufacture benign defaults.

## Step 3 — Output

- **When `total == 0`**: this branch is reachable only after the Step 2 query exited `0` and its payload parsed — a genuine empty result. Report "No tests found via static analysis. Try full suite or check with `grep -rn <symbol_name> tests/`."

+ **When `total == 0`**: reachable only after Step 2 exit `0` + parsed payload: genuine empty. Report "No tests found via static analysis. Try full suite or check with `grep -rn <symbol_name> tests/`."

**When `total > 0`**:

```

## Test impact: <qname>

**Affected tests** (<total> files, <via_call> via call/import graph, <via_mock> via mocks):

<test_files as bullet list>

**Run:**

```

<pytest_cmd>

```

**Caveat:** dynamic-dispatch / hook-callback callers are not in the static graph — <hint>.

</if>

```

- Output routing: if `total >= 5`, derive a free (non-colliding) path first, then write the report to the printed path:

+ Output routing: if `total >= 5`, derive free non-colliding path; write report there:

```bash

# timeout: 3000

BRANCH=$(git branch --show-current 2>/dev/null | tr '/' '-'); BRANCH="${BRANCH:-main}"

# never overwrite — a same-day re-run on another target would replace an unrelated report

TI_OUT=".temp/output-test-impact-${BRANCH}-$(date +%Y-%m-%d).md"; _n=1

while [ -e "$TI_OUT" ]; do _n=$((_n+1)); TI_OUT=".temp/output-test-impact-${BRANCH}-$(date +%Y-%m-%d)-${_n}.md"; done

printf '%s\n' "$TI_OUT"

```

</workflow>