acp-transcript-sync for Openclaw

To start using this capability within your Openclaw Skills workflow, follow these integration steps:

1. Locate the Script

Ensure you have the sync script available in your environment, typically located under:

<skill-dir>/scripts/sync-acp-to-main.py

2. Manual Command Execution

Run the sync tool directly using python, providing the mandatory sub-session details:

python3 scripts/sync-acp-to-main.py <acp_session_id> <acp_agent> [<main_session_id>] [<main_agent>]

Examples:

Explicitly defining both parent and child profiles:

python3 scripts/sync-acp-to-main.py f935963c-b8dd-431c-9d85-85e0e5a41341 cfuse 44c3f987-f9ad-4a02-b107-768ae9f517d3

Auto-detecting the active main session and parent agent:

python3 scripts/sync-acp-to-main.py f935963c-b8dd-431c-9d85-85e0e5a41341 cfuse

3. Verify the Synchronization

To manually confirm that transcripts have been successfully synchronized to the parent session, you can parse the tail end of the parent's session JSONL log:

tail -5 ~/.openclaw/agents/<main_agent>/sessions/<main_session_id>.jsonl | python3 -c "
import sys, json
for line in sys.stdin:
    d = json.loads(line)
    if d.get('type') == 'message':
        msg = d.get('message', {})
        content = msg.get('content', [])
        text = ''.join(c.get('text','') for c in content if c.get('type')=='text')
        print(f\"[{msg.get('role')}] {text[:100]}\")
"

Script Parameters

Transcript Merging & Formatting

To ensure distinct source tracking, merged entries are tagged with clear reasoning indicators based on the executing agent:

• [cfuse reasoning process] - Standard log entry for the cfuse runtime.
• [cfuse reasoning process] user - User-submitted prompt in the child session.
• [cfuse reasoning process] assistant - Model reasoning and outputs from the child agent.