From a0d2f6e4feb52fce6a28f22040eb6d471c97d225 Mon Sep 17 00:00:00 2001 From: Peter Steinberger Date: Sun, 1 Mar 2026 23:45:35 +0000 Subject: [PATCH] docs(changelog): note skills security hardening --- CHANGELOG.md | 1 + 1 file changed, 1 insertion(+) diff --git a/CHANGELOG.md b/CHANGELOG.md index fb084a87f..9cfe6b137 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -92,6 +92,7 @@ Docs: https://docs.openclaw.ai ### Fixes +- Security/Skills: harden skill installer metadata parsing by rejecting unsafe installer specs (brew/node/go/uv/download) and constrain plugin-declared skill directories to the plugin root (including symlink-escape checks), with regression coverage. - ACP/Harness thread spawn routing: force ACP harness thread creation through `sessions_spawn` (`runtime: "acp"`, `thread: true`) and explicitly forbid `message action=thread-create` for ACP harness requests, avoiding misrouted `Unknown channel` errors. (#30957) Thanks @dutifulbob. - Docs/ACP permissions: document the correct `permissionMode` default (`approve-reads`) and clarify non-interactive permission failure behavior/troubleshooting guidance. (#31044) Thanks @barronlroth. - Security/Logging utility hardening: remove `eval`-based command execution from `scripts/clawlog.sh`, switch to argv-safe command construction, and escape predicate literals for user-supplied search/category filters to block local command/predicate injection paths.