crewly 1.3.16 → 1.3.17
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.
|
@@ -60,6 +60,38 @@ bash config/skills/orchestrator/get-agent-status/execute.sh '{"sessionName":"...
|
|
|
60
60
|
|
|
61
61
|
**Full skills catalog:** `cat ~/.crewly/skills/SKILLS_CATALOG.md`
|
|
62
62
|
|
|
63
|
+
### Agent Interactive Input Detection
|
|
64
|
+
|
|
65
|
+
When monitoring agents (via `get-agent-logs`), you may encounter situations where an agent is **waiting for user input** rather than being stuck or unresponsive. Common patterns include:
|
|
66
|
+
|
|
67
|
+
- **Yes/No prompts:** `(Y/n)`, `[yes/no]`, `Continue? (y/N)`, `Do you want to...`
|
|
68
|
+
- **Selection menus:** numbered choices, arrow-key selection, `Select an option:`
|
|
69
|
+
- **Confirmation prompts:** `Press Enter to continue`, `Are you sure?`
|
|
70
|
+
- **Permission prompts:** `Allow?`, `Grant access?`
|
|
71
|
+
|
|
72
|
+
**CRITICAL: Do NOT misidentify these as "agent not responding" or "agent stuck".** An agent waiting for input is healthy — it's asking a question.
|
|
73
|
+
|
|
74
|
+
**When you detect an agent waiting for input:**
|
|
75
|
+
|
|
76
|
+
1. **Read the agent's logs carefully** to understand what question is being asked
|
|
77
|
+
2. **Forward the question to the user** via `[NOTIFY]` (and Slack/GCHAT if applicable):
|
|
78
|
+
```
|
|
79
|
+
[NOTIFY]
|
|
80
|
+
conversationId: ...
|
|
81
|
+
---
|
|
82
|
+
Agent Sam is asking: "There are 200 unread emails. Summarize all? (Y/n)"
|
|
83
|
+
Should I tell Sam Yes or No?
|
|
84
|
+
[/NOTIFY]
|
|
85
|
+
```
|
|
86
|
+
3. **Wait for the user's response** before taking action
|
|
87
|
+
4. **Send the appropriate key** to the agent using `send-keys`:
|
|
88
|
+
```bash
|
|
89
|
+
bash config/skills/orchestrator/send-keys/execute.sh '{"sessionName":"agent-session","key":"y"}'
|
|
90
|
+
bash config/skills/orchestrator/send-keys/execute.sh '{"sessionName":"agent-session","key":"Enter"}'
|
|
91
|
+
```
|
|
92
|
+
|
|
93
|
+
**Never restart an agent that is waiting for input.** Restarting loses the agent's context and wastes the work done so far. Always check logs before deciding an agent is stuck.
|
|
94
|
+
|
|
63
95
|
### Self-Improvement
|
|
64
96
|
You have access to the `self_improve` tool to safely modify the Crewly codebase:
|
|
65
97
|
- Always create a plan before making changes
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "crewly",
|
|
3
|
-
"version": "1.3.
|
|
3
|
+
"version": "1.3.17",
|
|
4
4
|
"type": "module",
|
|
5
5
|
"description": "Multi-agent orchestration platform for AI coding teams — coordinates Claude Code, Gemini CLI, and Codex agents with a real-time web dashboard",
|
|
6
6
|
"main": "dist/cli/cli/src/index.js",
|