openclaw-code-agent 4.1.1 → 4.1.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (5) hide show
  1. package/README.md +8 -6
  2. package/dist/index.js +129 -129
  3. package/openclaw.plugin.json +1 -1
  4. package/package.json +1 -1
  5. package/skills/code-agent-orchestration/SKILL.md +3 -1
package/openclaw.plugin.json CHANGED
@@ -2,7 +2,7 @@
2
2
  "id": "openclaw-code-agent",
3
3
  "name": "OpenClaw Code Agent",
4
4
  "description": "Multi-session coding-agent orchestration from OpenClaw chat",
5
- "version": "4.1.1",
5
+ "version": "4.1.2",
6
6
  "contracts": {
7
7
  "tools": [
8
8
  "agent_launch",
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "openclaw-code-agent",
3
- "version": "4.1.1",
3
+ "version": "4.1.2",
4
4
  "type": "module",
5
5
  "packageManager": "pnpm@10.30.0",
6
6
  "main": "dist/index.js",
package/skills/code-agent-orchestration/SKILL.md CHANGED
@@ -25,7 +25,7 @@ Use `openclaw-code-agent` to run Claude Code or Codex sessions as background cod
25
25
  - Set `workdir` to the target repo.
26
26
  - Use `permission_mode: "plan"` when the user wants a real review gate before implementation.
27
27
  - Use `permission_mode: "bypassPermissions"` only for autonomous execution.
28
- - `defaultWorktreeStrategy` now defaults to `off`. Opt into a worktree strategy explicitly when you want branch isolation.
28
+ - `defaultWorktreeStrategy` defaults to `delegate`, so new sessions normally use branch isolation and orchestrator-led follow-through. Use `worktree_strategy: "off"` only when the task must run in the main checkout or outside a git repo.
29
29
  - In `plan` mode, the plan belongs in normal session output. Do not ask the coding agent to write plan docs or transcript artifacts unless the user explicitly asked for a file.
30
30
 
31
31
  Example:
@@ -126,7 +126,9 @@ Use `permission_mode: "plan"` whenever the user wants a real planning checkpoint
126
126
 
127
127
  - Approval belongs to the user.
128
128
  - The plugin sends the canonical Approve / Revise / Reject prompt directly to the user.
129
+ - Telegram and Discord buttons use the shared direct-message presentation path; if buttons are missing, plain-text `Approve`, `Revise`, or `Reject` in the same thread drives the same decision path.
129
130
  - If the user requests changes, wait for the revised plan from that same session; the revised submission becomes the latest actionable review version automatically.
131
+ - If the user rejects the plan or the session is killed, treat older plan prompts as stale and verify state with `agent_sessions` before acting.
130
132
  - Wait for the user's answer, then forward it with `agent_respond(...)`.
131
133
  - Do not send a duplicate approval recap or second approval prompt.
132
134