feat(hooks-install): 技能同步到八支 hook 與重啟閘門的降級說法

What：`skills/hooks-install/SKILL.md` 四處增修。`description` 與目標段落的 hook 清單加入 `restart-gate.sh`，七支改八支；降級說明補上「四個 CLI 連部署後重啟閘門也接不上」與後果一句；第 4 步的冒煙說明補上重啟閘門的每條判定路徑；備註新增 `restart-gate.sh` 一條，並在 `session-timer.sh` 那一條補上「`start` 與 `restart` 會清除閘門」。

Why：技能是接線這件事的對外說法。hook 加了一支、技能還寫七支，回報就會少一項，而且降級說法不補會暗示每個 CLI 都擋得下來——這正是準則明文禁止的。`session-timer.sh` 那條也要補：清除閘門掛在那兩個事件上，接線少了它們，閘門會一路擋到使用者自己下逃生門。

How：降級那一段把後果講明白，不只說「接不上」：在那四個 CLI 上一次技能呼叫都擋不下來，狀態檔照樣寫、下一個工作階段開始照樣清，重啟本身只靠 `jsc-cli:deploy` 的收尾訊息。備註那一條寫出豁免清單與逃生門，並講明豁免的理由是異動報告與工作日誌要寫得完，讓接線的人知道哪些技能在閘門升起時仍然叫得動。

Who：`/jsc-hooks:hooks-install` 的接線流程與對使用者的回報。

skills/hooks-install/SKILL.md

@@ -1,15 +1,15 @@
 ---
 name: hooks-install
-description: Wire jsc hooks (STE100 guard, session timer, skill usage logger, SDLC model gate, plugin version guard, comment scope scanner, language guard) into every installed AI CLI, purging all pre-existing hooks first — third-party ones included, backed up before removal. Drive it per CLI through tools/wire-cli.sh purge, tools/wire-cli.sh, tools/wire-cli.sh smoke and tools/scan-hook-errors.sh. Hand any hook error, wiring or runtime, to jsc-hooks:repair, which must finish with a PR against develop; aborting the rest of the install to start that repair is allowed. Use after installing or updating the jsc plugin set; not for writing new hooks.
+description: Wire jsc hooks (STE100 guard, session timer, skill usage logger, SDLC model gate, plugin version guard, post-deploy restart gate, comment scope scanner, language guard) into every installed AI CLI, purging all pre-existing hooks first — third-party ones included, backed up before removal. Drive it per CLI through tools/wire-cli.sh purge, tools/wire-cli.sh, tools/wire-cli.sh smoke and tools/scan-hook-errors.sh. Hand any hook error, wiring or runtime, to jsc-hooks:repair, which must finish with a PR against develop; aborting the rest of the install to start that repair is allowed. Use after installing or updating the jsc plugin set; not for writing new hooks.
 ---
 
 # hooks-install — wire jsc hooks into every installed CLI
 
-Goal: make the seven hooks (`ste100-guard.sh`, `session-timer.sh`, `skill-usage.sh`, `sdlc-gate.sh`, `version-guard.sh`, `comment-scope.sh`, `lang-guard.sh`) effective in every CLI, with nothing else wired alongside them.
+Goal: make the eight hooks (`ste100-guard.sh`, `session-timer.sh`, `skill-usage.sh`, `sdlc-gate.sh`, `version-guard.sh`, `restart-gate.sh`, `comment-scope.sh`, `lang-guard.sh`) effective in every CLI, with nothing else wired alongside them.
 
 Install on a clean slate. Every CLI is purged of all hooks first, third-party ones included, so a later failure has exactly one owner. `tools/wire-cli.sh purge` backs up every file it touches before it removes anything, so the removal stays reversible.
 
-Only claude has PreToolUse, PostToolUse and UserPromptSubmit, so only claude reports `wired`. On codex, copilot, antigravity and kiro the version guard cannot be wired at all and the SDLC gate degrades to the skill-step check, so all four report `degraded` — report that gap as the script words it instead of implying every CLI is covered.
+Only claude has PreToolUse, PostToolUse and UserPromptSubmit, so only claude reports `wired`. On codex, copilot, antigravity and kiro neither the version guard nor the post-deploy restart gate can be wired at all, and the SDLC gate degrades to the skill-step check, so all four report `degraded` — report that gap as the script words it instead of implying every CLI is covered. On those four the restart gate blocks no skill call whatsoever: the state file is still written and still cleared at the next session start, so the restart itself rests on the `jsc-cli:deploy` closing message.
 
 `comment-scope.sh` and `lang-guard.sh` both reach all five, wired at the same set of places, but on a different event and at a different moment each. Report the timing per CLI; never state it as one uniform behaviour:
 
@@ -33,7 +33,7 @@ The detailed flow **MUST run as a sub agent**; the main agent only reports the s
 1. Run `jsc-cli/tools/detect-clis.sh`. Done when you hold the list of installed CLIs; when the list is empty, report that and stop.
 2. For each installed CLI, run `tools/wire-cli.sh purge {cli}`. The script backs up every file it touches, removes all hooks, re-reads each file to confirm the removal, and restores the backup by itself when a check fails. Done when every CLI has printed exactly one `status=purged|skipped|failed reason=...` line and you have noted the backup directory path from its `[jsc]` output.
 3. For each installed CLI, run `tools/wire-cli.sh {cli}`. The script owns both the wiring and its verification: it writes the config, alias or hook file inside a `<!-- jsc-hooks -->` (or `# jsc-hooks`) marker block, re-reads every file it wrote, and confirms the block is present and correctly placed before it prints a success status. Trust its first line, `status=wired|degraded|skipped|failed reason=...`. Exit 2 means a bad CLI name, not a wiring outcome — fix the name and rerun. Done when every installed CLI has printed exactly one `status=` line and none exited 2.
-4. For each installed CLI, run `tools/wire-cli.sh smoke {cli}`. This runs all seven hooks once each, every wired mode included, and catches what the wiring check cannot see: a hook that is wired correctly and still fails when it executes. Done when every CLI has printed one `status=ok|failed reason=...` line plus one result line per hook.
+4. For each installed CLI, run `tools/wire-cli.sh smoke {cli}`. This runs all eight hooks once each, every wired mode included, plus each decision path of the work-package check and of the restart gate, and catches what the wiring check cannot see: a hook that is wired correctly and still fails when it executes. Done when every CLI has printed one `status=ok|failed reason=...` line plus one result line per hook.
 5. For each installed CLI, run `tools/scan-hook-errors.sh --cli {cli}`. Only claude keeps hook results in its native records and can answer `clean` or `errors`; codex, copilot, antigravity and kiro answer `unavailable`, and their runtime evidence comes from step 4 alone. Done when every CLI has printed one `status=clean|errors|unavailable reason=...` line and the four `unavailable` CLIs are reported as exactly that, not as clean.
 6. For each error — `purge` failed, wiring failed, smoke failed, or a scanned error with `jsc=true` — run `tools/report-error.sh --hook {script name} --exit {code} --summary "{reason}" --cli {cli}` with the script's `[jsc]` output on stdin, then hand the failure to `jsc-hooks:repair`, which **MUST run as a sub agent** and must finish by opening a PR against `develop`. Aborting the remaining installs here is allowed as long as the repair starts. A scanned error with `jsc=false` belongs to a third-party hook: report it and leave it alone. Done when each error has either an `ERROR_{HASH}` page name on stdout, or an empty exit 0 meaning `JSC_WIKI_REPO_ERROR` and `JSC_WIKI_REPO` are both unset — in that second case carry the reason into step 7 instead. Skip this step when every CLI passed all four checks.
 7. Report four results per CLI — purge, wiring, smoke, scan — each with the reason its script printed, plus any `ERROR_{HASH}` page name and repair PR URL. Done when every detected CLI has exactly one status per check and every repair has a PR against `develop`.
@@ -41,7 +41,8 @@ The detailed flow **MUST run as a sub agent**; the main agent only reports the s
 ## Notes
 
 - Every hook script accepts both stdin JSON and environment variables (`JSC_CLI`, `JSC_SESSION_ID`, `JSC_SKILL`, `JSC_TOOL_NAME`, `JSC_MODEL`); `jsc-wrap.sh` sets the first two itself.
-- `session-timer.sh` takes `start` (keep an existing start time), `restart` (always overwrite it, for a CLI with no session id — kiro), `mark` and `report`. `wire-cli.sh` picks the right one per CLI; do not hand-edit the generated hook files.
+- `session-timer.sh` takes `start` (keep an existing start time), `restart` (always overwrite it, for a CLI with no session id — kiro), `mark` and `report`. `wire-cli.sh` picks the right one per CLI; do not hand-edit the generated hook files. `start` and `restart` also clear the restart gate whenever they decide this SessionStart is a new session, so the wiring of those two events is what lowers the gate after a restart — a CLI wired without them keeps the gate up until the user sets `JSC_RESTART_GATE=off`.
+- `restart-gate.sh` blocks jsc skill calls while `$JSC_HOME/restart-required` exists, so a freshly deployed skill set is not used by a process still running the old one. `jsc-cli:deploy` writes that file through `restart-gate.sh require {install|update} [{domain}...]` at the end of an install or update. Exempt skills stay callable — `jsc-cli:deploy`, `jsc-hooks:hooks-install`, `jsc-gitea:wiki`, `jsc-log:worklog`, `jsc-log:learn`, `jsc-meta:*` — because the change report and the worklog still have to be finished after a deploy. Escape hatch: `JSC_RESTART_GATE=off`.
 - `purge` reaches the user-level config only. Hooks that another plugin ships in its own `hooks.json` stay active, and uninstalling that plugin is the only way to clear them — say so when reporting, and treat their errors as third-party.
 - Backups land in `$JSC_HOME/backup/hooks/{cli}/{yyyyMMdd_HHmmss}/`, one directory per purge run, under the original file names. Hand that path to the user whenever a purge removed something.
 - `smoke` treats `sdlc-gate.sh check` exit 2 as healthy: that exit is the stage lock blocking a turn on purpose, not a runtime error. `comment-scope.sh` and `lang-guard.sh` exit 2 count as healthy for the same reason — the scan found something and warned about it. Their no-argument mode has no file name during smoke and exits 0 in silence; `sweep` depends on the worktree it runs in, so it answers 2 whenever that worktree happens to carry an offending comment, a simplified character or a mojibake sequence. None of these is a broken hook.