Update agent setup

.agents/skills/compile/SKILL.md

@@ -16,7 +16,7 @@ Presets: `gcc-` / `clang-` + `debug` (coverage), `release`, `asan` (ASan + UBSan
 
 ## What CI builds on every PR
 
-1. `clang-format` **19** over the tree (gates everything). Check `clang-format --version`: if it is not 19, ask the user to install it; a format diff seen only in CI points to the version gap.
+1. `clang-format` at the version pinned in `build.yml` over the tree (gates everything). Check `clang-format --version`: if it differs, ask the user to install it; a format diff seen only in CI points to the version gap.
 2. gcc and clang x amd64 and arm64, in `joinframework/join-ci`: `release` then `debug`; `asan` on amd64.
 3. CodeQL on `gcc-debug`.
 4. `gcovr` coverage on amd64 (gcc and clang, kept as artifacts); only gcc's is uploaded to Coveralls/Codecov/Codacy by `deploy-coverage`, which can fail the run.

.agents/skills/conformance/SKILL.md

@@ -7,17 +7,17 @@ description: Check dtoa/atod correctness versus main with the dconv-benchmark co
 
 Required at closing when conversion code changes (`dconv/include/dconv/`). No timing: any machine works. For speed see skill `performance`.
 
-Clone `https://github.com/joinframework/dconv-benchmark.git` into `<bench>` under `<scratchpad>`, never into the dconv tree. It pins dconv to a release tag, so point both builds at local checkouts, `origin/main` in a worktree:
+`<bench>` is `dconv-benchmark` next to dconv (`../dconv-benchmark`); if missing, clone `https://github.com/joinframework/dconv-benchmark.git` there, never into the dconv tree. It pins dconv to a release tag, so point both builds at local checkouts, `origin/main` in a worktree:
 
 ```bash
 git -C <dconv> fetch origin main && git -C <dconv> worktree add --detach <scratchpad>/base origin/main
-cmake -S <bench> -B <bench>/build/main   -DCMAKE_BUILD_TYPE=Release -DFETCHCONTENT_SOURCE_DIR_DCONV=<scratchpad>/base
-cmake -S <bench> -B <bench>/build/branch -DCMAKE_BUILD_TYPE=Release -DFETCHCONTENT_SOURCE_DIR_DCONV=<dconv>
+cmake -S <bench> -B <scratchpad>/build/main   -DCMAKE_BUILD_TYPE=Release -DFETCHCONTENT_SOURCE_DIR_DCONV=<scratchpad>/base
+cmake -S <bench> -B <scratchpad>/build/branch -DCMAKE_BUILD_TYPE=Release -DFETCHCONTENT_SOURCE_DIR_DCONV=<dconv>
 cmake --build <build> --target conformance
 <build>/benchmarks/conformance/conformance --output <scratchpad>/conformance-<main|branch>.json
 ```
 
-Remove the worktree at the end (`git worktree remove --force`), even on failure.
+Check each tree measures the intended dconv: `grep FETCHCONTENT_SOURCE_DIR_DCONV <build>/CMakeCache.txt` must show its path, otherwise the numbers are the pinned tag. Remove the worktree at the end (`git worktree remove --force`), even on failure.
 
 ## Pass criteria
 

.agents/skills/coverage/SKILL.md

@@ -14,19 +14,19 @@ cmake --preset $C-debug && cmake --build --preset $C-debug
 find build/$C/debug -name '*.gcda' -delete
 ctest --preset $C-debug
 gcovr --root . --gcov-executable "$GCOV" --exclude-pattern-prefix LCOV --exclude-branches-by-pattern '.*' \
-      --exclude '.*/tests/.*' --exclude 'tools/.*' --json-summary <scratchpad>/$C-after.json --print-summary build/$C/debug
+      --exclude '.*/tests/.*' --exclude 'tools/.*' --json-summary <scratchpad>/$C-${SIDE:-after}.json --print-summary build/$C/debug
 ```
 
-Baseline: same steps on `origin/main` in a worktree (`git fetch origin main && git worktree add --detach <scratchpad>/base origin/main`) into `$C-before.json`; remove it at the end (`git worktree remove --force`), even on failure.
+Baseline: same steps on `origin/main` in a worktree (`git fetch origin main && git worktree add --detach <scratchpad>/base origin/main`) with `SIDE=before`, into `$C-before.json`; remove it at the end (`git worktree remove --force`), even on failure.
 
 Compare the total and every changed header. Any drop fails: add tests (or `LCOV_EXCL_*` for truly unreachable code, sparingly) and re-measure.
 
-Needs `gcovr` and `llvm-cov` matching clang. Without them locally, save the steps above to `<scratchpad>/coverage.sh` and run it in CI's image (`docker pull joinframework/join-ci:latest` first if it lacks `gcovr`), on copies so the host `build/` is untouched:
+Needs `gcovr` and `llvm-cov` matching clang. Without them locally, save the steps above, after `set -e`, to `<scratchpad>/coverage.sh` and run it in CI's image (`docker pull joinframework/join-ci:latest` first if it lacks `gcovr`), on copies so the host `build/` is untouched:
 
 ```bash
 rsync -a --exclude build ./ <scratchpad>/head/
 docker run --rm --user "$(id -u):$(id -g)" -v <scratchpad>:<scratchpad> -w <scratchpad>/head \
        joinframework/join-ci:latest bash <scratchpad>/coverage.sh
 ```
 
-Same for the baseline with `-w <scratchpad>/base`. If neither works, report coverage as **not verified**, never "not degraded".
+Same for the baseline with `-e SIDE=before -w <scratchpad>/base`. If neither works, report coverage as **not verified**, never "not degraded".

.agents/skills/create-pr/SKILL.md

@@ -15,7 +15,7 @@ description: Branch, commit, push and open a PR with the dconv conventions. Use
    - `hotfix/`: small quick fix, minimal diff.
 
    Ask once if bugfix vs hotfix is unclear.
-4. **Version**: bump `VERSION` in every PR: minor for a feature or an API change, patch otherwise (bugfix, hotfix, tools, CI, docs, agent config), major only on request. Minor or major also sets `PROJECT_NUMBER` = `<major>.<minor>` in `doxyfile`.
+4. **Version**: bump `VERSION` in every PR, unless it is already ahead of the latest tag at an equal or higher level: minor for a feature or an API change, patch otherwise (bugfix, hotfix, tools, CI, docs, agent config), major only on request. Minor or major also sets `PROJECT_NUMBER` = `<major>.<minor>` in `doxyfile`.
 5. **Commit**: explicit paths (no `git add -A`), one commit. Message: one capitalized imperative line, about 50 chars, no period, no body, no PR number, no trailer: `Stop installing tool headers`, or `Schubfach algorithm` for a new algorithm.
 6. **Push**: `git push -u origin <branch>`. Never rewrite pushed history; report a rejected push.
 7. **PR**: title = commit line, no body. Ask first; with `gh`: `gh pr create --base main --title "<Title>" --body ""`. Otherwise give `https://github.com/joinframework/dconv/compare/main...<branch>?expand=1`.

.agents/skills/performance/SKILL.md

@@ -9,7 +9,7 @@ Required at closing when conversion code changes (`dconv/include/dconv/`). Done
 
 ## 1. Ask first
 
-Ask the user: **is a real machine available (ideally x86_64 and aarch64)?** Clone `dconv-benchmark` there as in skill `conformance`, never into the dconv tree.
+Ask the user: **is a real machine available (ideally x86_64 and aarch64)?** Use the `dconv-benchmark` checkout of skill `conformance`.
 
 ## 2. Performance mode
 
@@ -25,10 +25,10 @@ Use `sudo` only when justified; otherwise give the commands with the `! ` prefix
 
 ## 3. Compare
 
-Same `build/main` / `build/branch` setup as skill `conformance` (local checkouts via `FETCHCONTENT_SOURCE_DIR_DCONV`, `main` in a worktree). Build `bench_dtoa_dconv` and `bench_atod_dconv` (+ `bench_dtoa_dconvgrisu{,2,3}` if Grisu changed).
+Same `<scratchpad>/build/main` / `<scratchpad>/build/branch` setup as skill `conformance` (local checkouts via `FETCHCONTENT_SOURCE_DIR_DCONV`, `main` in a worktree). Build `bench_dtoa_dconv` and `bench_atod_dconv` (+ `bench_dtoa_dconvgrisu{,2,3}` if Grisu changed).
 
 Binaries: `<build>/benchmarks/performance/bench_*`. Export the same `GLIBC_TUNABLES` as `scripts/run_bench.sh`. Run them alternately main/branch, several times, `taskset -c 2`, with `--benchmark_min_warmup_time=0.1 --benchmark_min_time=0.2s`; keep the minimum. Datasets: `Canada`, `Random`, `Unit`, `Integer`, `Digits01..17`. Remove the worktree at the end, even on failure.
 
 ## 4. Report
 
-Machine (CPU, arch, governor/boost, native or WSL2), repetitions, and a before/after table per dataset with the relative change. A change is real only beyond the spread; never claim an unmeasured gain.
+Machine (CPU, arch, governor/boost, native or WSL2), repetitions, and a before/after table per dataset with the relative change. A change is real only beyond the spread; never claim an unmeasured gain. When a delta is close to the spread, rerun with 5 repetitions per side and decide with google benchmark's `compare.py` (`<build>/_deps/googlebenchmark-src/tools/compare.py benchmarks <main>.json <branch>.json`, needs `scipy`; merge each side's runs into one JSON first): with 3 per side its Mann-Whitney test cannot reach p < 0.05. A mechanism read from the diff is a hypothesis until `perf stat -e cycles,instructions,cache-misses,branch-misses` shows it (needs `kernel.perf_event_paranoid` <= 1).

.agents/skills/release/SKILL.md

@@ -20,3 +20,4 @@ description: Tag and publish a dconv release from main. Use when asked to releas
 5. **Downstream**: list the updates, each done in its own repo with its own PR, only if the user asks:
    - join: `find_package(dconv <version> REQUIRED)` in `data/CMakeLists.txt`, `find_dependency(dconv <version>)` in `cmake/joinConfig.cmake.in`, `libdconv-dev (>= <version>)` in `cmake/packaging.cmake`, the `.deb` URL and file name in `README.md`.
    - dconv-benchmark: `GIT_TAG v<version>` in `cmake/deps.cmake`, the dconv rows of the README libraries table (results and plots only after a new benchmark run).
+   - json-benchmark, pack-benchmark: `GIT_TAG v<version>` in `cmake/deps.cmake`.

.claude/hooks/format.sh

@@ -1,17 +1,17 @@
 #!/usr/bin/env bash
-missing () {
-    echo "format hook: $1 is missing, nothing was formatted. Ask the user to install it (AGENTS.md, Definition of done)." >&2
-    exit 2
-}
-input=$(cat)
-case "$input" in *'.hpp"'*|*'.cpp"'*) ;; *) exit 0 ;; esac
-command -v jq >/dev/null || missing jq
-file=$(jq -r '.tool_input.file_path // empty' <<<"$input")
-case "$file" in
-    *.hpp|*.cpp)
-        command -v clang-format >/dev/null || missing clang-format
-        [ -f "$file" ] || exit 0
-        clang-format -i "$file" || { echo "format hook: clang-format failed on $file" >&2; exit 2; }
-        ;;
-esac
+re='"file_path": *"([^"]*\.(hpp|cpp))"'
+sre='"session_id": *"([^"]*)"'
+in=$(cat)
+[[ $in =~ $sre ]] && mark=${TMPDIR:-/tmp}/format-hook-${BASH_REMATCH[1]}
+[[ $in =~ $re ]] || exit 0
+file=${BASH_REMATCH[1]}
+[[ $file == "$CLAUDE_PROJECT_DIR"/* ]] || exit 0
+v=$(sed -n "s/.*clang-format-version: *'\([0-9]*\)'.*/\1/p" "$CLAUDE_PROJECT_DIR"/.github/workflows/build.yml | head -1)
+cf=clang-format-$v
+command -v "$cf" >/dev/null || cf=clang-format
+command -v "$cf" >/dev/null || { echo "format hook: clang-format is missing, nothing was formatted. Ask the user to install it (AGENTS.md, Definition of done)." >&2; exit 2; }
+[ -f "$file" ] || exit 0
+"$cf" -i "$file" || { echo "format hook: $cf failed on $file" >&2; exit 2; }
+have=$("$cf" --version | sed -n 's/.*version \([0-9]*\)\..*/\1/p')
+[ -z "$v" ] || [ "$have" = "$v" ] || [ -e "$mark" ] || { touch "$mark"; echo "{\"hookSpecificOutput\":{\"hookEventName\":\"PostToolUse\",\"additionalContext\":\"format hook: formatted with clang-format $have, CI uses $v; ask the user to install clang-format $v.\"}}"; }
 exit 0

.claude/rules/cpp-style.md

@@ -8,7 +8,7 @@ paths:
 
 # C++ style
 
-`.clang-format` is enforced (hook, or `clang-format` 19); never hand-format.
+`.clang-format` is enforced (hook, or `clang-format` at the CI version); never hand-format.
 
 ## Layout
 - C++20, header-only, no language extensions; GNU builtins and attributes already in use (`__builtin_expect`, `__attribute__ ((cold, noinline))`) are accepted. Space before every `(` (`foo (a)`, `TEST (s, n)`); Allman braces on every control statement.

AGENTS.md

@@ -18,7 +18,7 @@ cmake --preset gcc-debug && cmake --build --preset gcc-debug && ctest --preset g
 
 ## Agent files
 
-Skills in `.agents/skills/<name>/SKILL.md`: compile, test, coverage, conformance, performance, add-gtest, gen-tables, create-pr, release; `<scratchpad>` in them is a temporary directory outside the repository. Run `clang-format` (CI uses 19) on changed C++ files.
+Skills in `.agents/skills/<name>/SKILL.md`: compile, test, coverage, conformance, performance, add-gtest, gen-tables, create-pr, release; `<scratchpad>` in them is a temporary directory outside the repository. Run `clang-format` on changed C++ files at the version pinned in `.github/workflows/build.yml`.
 
 Changes to `AGENTS.md` and `.agents/` keep them tool-agnostic and short; `.claude/` is Claude Code only.
 

VERSION

@@ -1 +1 @@
-1.2.2
+1.2.3