fix(exec): restore runtime-aware implicit host default

docs/gateway/security/index.md

@@ -191,7 +191,7 @@ If more than one person can DM your bot:
 - **Local disk hygiene** (permissions, symlinks, config includes, “synced folder” paths).
 - **Plugins** (extensions exist without an explicit allowlist).
 - **Policy drift/misconfig** (sandbox docker settings configured but sandbox mode off; ineffective `gateway.nodes.denyCommands` patterns because matching is exact command-name only (for example `system.run`) and does not inspect shell text; dangerous `gateway.nodes.allowCommands` entries; global `tools.profile="minimal"` overridden by per-agent profiles; extension plugin tools reachable under permissive tool policy).
-- **Runtime expectation drift** (for example `tools.exec.host="sandbox"` while sandbox mode is off, which now fails closed because no sandbox runtime is available).
+- **Runtime expectation drift** (for example `tools.exec.host="sandbox"` while sandbox mode is off, which fails closed because no sandbox runtime is available).
 - **Model hygiene** (warn when configured models look legacy; not a hard block).
 
 If you run `--deep`, OpenClaw also attempts a best-effort live Gateway probe.
@@ -534,7 +534,7 @@ Even with strong system prompts, **prompt injection is not solved**. System prom
 - Prefer mention gating in groups; avoid “always-on” bots in public rooms.
 - Treat links, attachments, and pasted instructions as hostile by default.
 - Run sensitive tool execution in a sandbox; keep secrets out of the agent’s reachable filesystem.
-- Note: sandboxing is opt-in. If sandbox mode is off, `host=sandbox` fails closed even though tools.exec.host defaults to sandbox. To run on the gateway host, set `host=gateway` and configure exec approvals.
+- Note: sandboxing is opt-in. If sandbox mode is off, explicit `host=sandbox` fails closed because no sandbox runtime is available. Implicit exec still runs on the gateway host; set `host=gateway` if you want that behavior to be explicit in config.
 - Limit high-risk tools (`exec`, `browser`, `web_fetch`, `web_search`) to trusted agents or explicit allowlists.
 - If you allowlist interpreters (`python`, `node`, `ruby`, `perl`, `php`, `lua`, `osascript`), enable `tools.exec.strictInlineEval` so inline eval forms still need explicit approval.
 - **Model choice matters:** older/smaller/legacy models are significantly less robust against prompt injection and tool misuse. For tool-enabled agents, use the strongest latest-generation, instruction-hardened model available.

docs/tools/exec.md

@@ -29,7 +29,7 @@ Background sessions are scoped per agent; `process` only sees sessions from the
 
 Notes:
 
-- `host` defaults to `sandbox`.
+- `host` defaults to `sandbox` when sandbox runtime is active for the session; otherwise it defaults to `gateway`.
 - `elevated` forces `host=gateway`; it is only available when elevated access is enabled for the current session/provider.
 - `gateway`/`node` approvals are controlled by `~/.openclaw/exec-approvals.json`.
 - `node` requires a paired node (companion app or headless node host).
@@ -52,7 +52,7 @@ Notes:
 
 - `tools.exec.notifyOnExit` (default: true): when true, backgrounded exec sessions enqueue a system event and request a heartbeat on exit.
 - `tools.exec.approvalRunningNoticeMs` (default: 10000): emit a single “running” notice when an approval-gated exec runs longer than this (0 disables).
-- `tools.exec.host` (default: `sandbox`)
+- `tools.exec.host` (default: runtime-aware: `sandbox` when sandbox runtime is active, `gateway` otherwise)
 - `tools.exec.security` (default: `deny` for sandbox, `allowlist` for gateway + node when unset)
 - `tools.exec.ask` (default: `on-miss`)
 - `tools.exec.node` (default: unset)