claudemd-cli 0.44.0 → 0.46.0

package/CHANGELOG.md

@@ -8,6 +8,28 @@ All notable changes to the `claudemd` plugin. This changelog tracks plugin artif
 - **Canonical spec version source**: `spec/CLAUDE.md` top-line title (`# AI-CODING-SPEC vX.Y.Z — Core`) + `spec/CLAUDE-changelog.md` top `##` entry.
 - **Plugin semver vs spec semver** are independent: plugin patch (0.2.0 → 0.2.1) may ship when spec is unchanged (this release); plugin minor (0.1.9 → 0.2.0) ships when spec minor updates (v0.2.0 shipped spec v6.10.0).
 
+## [0.46.0] - 2026-07-13
+
+**Minor — §8 detection-precision batch: close command-grouping/substitution false-negatives (SEC-3) + recognize mktemp-provenance temp cleanup (SEC-4).** Hook-code only; spec text unchanged (stays v6.19.0). From the 2026-07-13 comprehensive audit + the §8 detection-precision design review (`docs/s8-detection-precision-design-2026-07-13.md`). All changes sandbox-verified against the full corpus + adversarial FN-matrix (`DISABLE_RULE_HITS_LOG=1`) before landing.
+
+- **SEC-3 — command-grouping/substitution FN closure** (`hooks/pre-bash-safety-check.sh`): a command inside `$(…)` / `(…)` / `{…;}` / backtick **executes at eval time** but bypassed all three §8 gates. Fixes: the rm segment splitter now breaks on `(` `)` backtick (NOT `{}` — that is brace-`${VAR}` far more often, splitting it corrupted the var target) + strips a leading group opener so `(rm`/`{ rm` is seen as rm + strips grouping chars from the residue; the curl sink tolerates a closing `)`/`}`; the npx anchor gains `(`/`{` so `$(npx …)` is caught. `\npx`/`env npx` wrapper coverage and `${VAR}` targets are preserved (233-case corpus zero-regression).
+- **SEC-4 — mktemp-provenance recognition** (`hooks/pre-bash-safety-check.sh`): `T=$(mktemp -d); … rm -rf "$T"` (the §8.V4 sandbox-disposal idiom, previously denied as an unvalidated var) is now allowed when the target var is assigned **in the same command** from `$(mktemp …)` / `` `mktemp …` `` — mktemp output is a fresh, uniquely-named path. Additive recognition analogous to the existing `${VAR:?}` guard; new telemetry event `rm-rf-allow-provenance`. Cross-command mktemp, fake `mktemp` (`$(echo mktemp)`), transitive vars (`$S/x`), and `..` escapes stay strict.
+- **Test completeness** (`tests/scripts/hard-rules-drift.test.js`): the hard-rules-8 reverse-completeness assertion now enumerates **both** deny idioms (`HIT_SECTIONS+=` and direct `hook_record <hook> <deny-verb> … '§…'`); the prior regex saw only the first, leaving `§10-V` / `§11-memory-read` / `§7-ship-baseline` invisible to the demote-accounting guard.
+- **Corpus** (`tests/fixtures/bash-safety/corpus.tsv`): +22 rows (F7 grouping/substitution deny + FP controls, F8 mktemp-provenance allow + escape/fake/cross-call deny). Suite 233 → 255.
+- **Docs** (`docs/RULE-HITS-SCHEMA.md`): document `rm-rf-allow-provenance` event + section mapping.
+- **Deferred**: the npx command-position anchor rework (fix npx-plain-words-in-quoted-commit-message FP without regressing `env npx` wrapper coverage) needs a wrapper-aware token walk — see `tasks/s8-npx-message-fp-deferred.md`.
+- **Tests**: full `npm test` green (corpus 255/255, node --test all pass, integration lifecycle PASS); shellcheck clean; positioning as guardrail (env/bypass-token still bypass) unchanged.
+
+## [0.45.0] - 2026-07-13
+
+**Minor — ships spec v6.19.0: §2.2 Runbook fast-path (ship-time token-efficiency relaxation).** When extended would load solely because of ship/release (incl. released-artifact L3) and the project's ship-runbook memory carries a current-version coverage stamp (`covers: §EXT §12 … @ v<spec>`), the ship reads the runbook + targeted-reads the stamped §EXT sections instead of the full ~47KB file. Stamp missing/stale → full load + stamp refresh (self-healing, one full re-read per spec release). All §12 HARD obligations bind unchanged. Companion to the v6.16.0 ship-runbook memory consolidation — the two together make a routine ship cost one runbook Read + ~7KB of targeted spec reads.
+
+- **Spec** (`spec/CLAUDE.md` §2.2 + §2.1, `spec/CLAUDE-extended.md` §12 + new §2.1-EXT + Recent changes + Sizing, `spec/CLAUDE-changelog.md`): v6.18.0 → v6.19.0. Paired net-delete per §0.1: **C4 consumed** — §2.1 Model-tiering Sonnet/Opus category enumeration moved to new §EXT §2.1-EXT; safety invariants (inherit default / NEVER-downgrade list / verifier ≥ generator / anomalous re-run / evidence bar) stay in core. Candidate pool now empty.
+- **Manifest** (`spec/hard-rules.json`): `spec_version` v6.18.0 → v6.19.0 (no rule add/remove — fast-path bounds live inside the existing `§12-ship-pipeline-hardening` HARD block; anchor unchanged).
+- **Descriptions** (`.claude-plugin/plugin.json`, `.claude-plugin/marketplace.json` ×2, `README.md` ×3): spec family tag v6.18 → v6.19.
+- **Memory (project-side, not in repo)**: claudemd ship runbook (`feedback_claudemd_ship_from_main_atomic.md`) gains the coverage stamp `covers: §EXT §12, §EXT §13 META, §EXT §2-EXT released-artifact checklist @ v6.19.0`.
+- **Tests**: full `npm test` green; `version-cascade-check` v6.19 consistent + Sizing within ±20B; shellcheck + bash-3.2 gate clean.
+
 ## [0.44.0] - 2026-07-13
 
 **Minor — ships spec v6.18.0: §1 Language-contract refinement.** Reasoning/思考 moves from the user's-language bucket to English (joining code / comments / commits); a docs split is added — local analysis/audit docs follow the user's language, shipped reference/contract docs (ARCHITECTURE / HOOK-PROTOCOL / RULE-HITS-SCHEMA / ADDING-NEW-HOOK / cross-project-pilot) stay English for adopters; Done narrative made explicit in the user's-language bucket. Code artifacts / CHANGELOG / PR / log-strings / config-keys / CLI-labels unchanged (English). Operator-requested with boundaries confirmed before edit.

package/README.md

@@ -1,12 +1,12 @@
 # claudemd
 
-> A **personal AI-coding discipline harness**: one developer's opinionated **AI-CODING-SPEC v6.18**, encoded as Claude Code shell hooks and shipped with the plugin. Built and dogfooded on my own repos — fork and adapt, don't adopt wholesale.
+> A **personal AI-coding discipline harness**: one developer's opinionated **AI-CODING-SPEC v6.19**, encoded as Claude Code shell hooks and shipped with the plugin. Built and dogfooded on my own repos — fork and adapt, don't adopt wholesale.
 
 [![CI](https://github.com/sdsrss/claudemd/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/sdsrss/claudemd/actions/workflows/ci.yml)
 [![npm](https://img.shields.io/npm/v/claudemd-cli.svg)](https://www.npmjs.com/package/claudemd-cli)
 [![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
 
-claudemd plugs into the Claude Code hook system to **block commits, pushes, and bash commands** that violate AI-CODING-SPEC v6.18 — banned vocabulary in commit messages, `rm -rf $VAR` without variable validation, ship-on-red-CI, unread `MEMORY.md` entries during release flows, and more. The spec itself (`CLAUDE.md` + `CLAUDE-extended.md` + `CLAUDE-changelog.md` + `OPERATOR.md`) ships with the plugin and installs into `~/.claude/`, so the rules Claude Code reads at session start match the rules the hooks enforce. (`OPERATOR.md` is the human-only spec-maintenance handbook — Agent-loaded files are the CLAUDE trio.)
+claudemd plugs into the Claude Code hook system to **block commits, pushes, and bash commands** that violate AI-CODING-SPEC v6.19 — banned vocabulary in commit messages, `rm -rf $VAR` without variable validation, ship-on-red-CI, unread `MEMORY.md` entries during release flows, and more. The spec itself (`CLAUDE.md` + `CLAUDE-extended.md` + `CLAUDE-changelog.md` + `OPERATOR.md`) ships with the plugin and installs into `~/.claude/`, so the rules Claude Code reads at session start match the rules the hooks enforce. (`OPERATOR.md` is the human-only spec-maintenance handbook — Agent-loaded files are the CLAUDE trio.)
 
 A standalone CLI (`npx claudemd-cli`) reuses the same `banned-vocab.patterns` source for git pre-commit hooks, GitHub Actions, and other agents that don't run inside Claude Code.
 
@@ -60,7 +60,7 @@ Verify in one command (Linux): `node --version && jq --version && gh --version &
 | 16 shell hooks | `banned-vocab-check` · `pre-bash-safety-check` · `ship-baseline-check` · `residue-audit` · `memory-read-check` · `memory-prompt-hint` · `mid-spine-yield-scan` · `sandbox-disposal-check` · `session-start-check` · `session-extended-read` · `session-summary` · `session-end-check` · `transcript-vocab-scan` · `transcript-structure-scan` · `version-sync` · `mem-audit` |
 | 15 slash commands | `/claudemd-install` · `/claudemd-status` · `/claudemd-update` · `/claudemd-audit` · `/claudemd-toggle` · `/claudemd-doctor` · `/claudemd-analyze` · `/claudemd-uninstall` · `/claudemd-rules` · `/claudemd-clean-residue` · `/claudemd-sparkline` · `/claudemd-sampling-audit` · `/claudemd-bypass-audit` · `/claudemd-design-adopt` · `/claudemd-statusline` |
 | 1 standalone CLI | `claudemd-cli lint` · `claudemd-cli audit` ([npm: `claudemd-cli`](https://www.npmjs.com/package/claudemd-cli)) |
-| Spec v6.18 | `~/.claude/CLAUDE.md` · `CLAUDE-extended.md` · `CLAUDE-changelog.md` · `OPERATOR.md` (backup-before-overwrite) |
+| Spec v6.19 | `~/.claude/CLAUDE.md` · `CLAUDE-extended.md` · `CLAUDE-changelog.md` · `OPERATOR.md` (backup-before-overwrite) |
 | StatusLine (opt-out) | PS1-style line — `user@host:dir (branch) Model [ctx:N% · 5h:N% · 7d:N%]` (`dir` = cwd basename; context / 5-hour quota / weekly quota, all **used %**, read from Claude Code's `rate_limits` payload; quota segments auto-hide when the data is absent, or force-hide with `DISABLE_STATUSLINE_QUOTA=1`) — wired into `~/.claude/settings.json` on install **only when the slot is empty**; an existing statusline is left untouched. Skip entirely with `CLAUDEMD_NO_STATUSLINE=1`. Manage via `/claudemd-statusline`. |
 
 Install backs up a hand-written `~/.claude/CLAUDE.md` (any file without the `# AI-CODING-SPEC` H1) to `~/.claude/backup-<ISO>/` before overwriting (last 5 kept automatically). An already-installed claudemd spec is overwritten **without** a backup — deliberate (v0.23.11): the sole backup is always your own content, so `restore` can never return a stale spec instead. Uninstall offers `keep / restore / delete`; `delete` requires an extra confirmation.
@@ -338,7 +338,7 @@ claudemd/
 ├── commands/                 # 15 slash-command markdown files
 ├── bin/                      # standalone CLI entrypoint (claudemd-lint.js → `npx claudemd-cli` on npmjs.org)
 ├── scripts/                  # 18 Node.js scripts + scripts/lib/ (single-source registry, lint, etc.)
-├── spec/                     # shipped v6.18.0 CLAUDE*.md trio + OPERATOR.md + hard-rules.json manifest
+├── spec/                     # shipped v6.19.0 CLAUDE*.md trio + OPERATOR.md + hard-rules.json manifest
 ├── tests/                    # hook shell tests + Node.js tests + integration + fixtures
 ├── docs/                     # ADDING-NEW-HOOK.md + RULE-HITS-SCHEMA.md + superpowers/
 └── .github/workflows/        # ci.yml (ubuntu+macOS × node 20) + npm-publish.yml (tag-triggered)

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claudemd-cli",
-  "version": "0.44.0",
+  "version": "0.46.0",
   "description": "Standalone CLI for §10-V banned-vocab + transcript scanning. Companion to the claudemd Claude Code plugin (github.com/sdsrss/claudemd) for use in git pre-commit hooks, GitHub Actions, and other agents.",
   "type": "module",
   "bin": {