Conversation
Documents ACPAgent usage, configuration, metrics, and cleanup with a ready-to-run example referencing examples/01_standalone_sdk/40_acp_agent_example.py. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
all-hands-bot
left a comment
all-hands-bot
left a comment
There was a problem hiding this comment.
🟡 Acceptable - Documentation is clear and practical, but has organizational issues and unverified syntax.
Key Issue: Important information about ACP delegation model is scattered. The fact that this replaces direct LLM calls (no API key needed, server manages everything) should be front and center, not discovered at line 143.
|
@OpenHands read the PR and apply all suggestions |
|
I'm on it! simonrosenberg can track my progress at all-hands.dev |
- Replace 'wrap focus={5-7}' with 'highlight={5-7}' per Mintlify standards
- Remove 'wrap' attribute from other code blocks
- Add Note callout explaining key differentiator (no LLM_API_KEY needed)
- Move 'What ACPAgent Does Not Support' section after Basic Usage
- Expand permission auto-approval explanation with security context
Co-authored-by: openhands <openhands@all-hands.dev>
SummaryI've successfully applied all review suggestions from PR #351. The changes have already been committed and pushed to the Checklist of Review Suggestions Applied:
Commit Details:
The changes are concise and focused solely on addressing the review feedback. No extraneous modifications were made. |
all-hands-bot
left a comment
all-hands-bot
left a comment
There was a problem hiding this comment.
🟡 Acceptable - Good improvements from previous feedback, but missing the "how to run" instructions that users need.
Key Issue: You import RunExampleCode but never use it, and provide no alternative bash block. Users reading this will see example code but no clear command to execute it.
|
|
||
| ## Basic Usage | ||
|
|
||
| ```python icon="python" highlight={5-7} |
There was a problem hiding this comment.
🟡 Suggestion: highlight={5-7} includes a blank line (line 6). Either highlight just the ACPAgent instantiation with highlight={5} or the full pattern with highlight={5,7-9} (skipping the blank).
| ``` | ||
|
|
There was a problem hiding this comment.
🟠 Important: You imported <RunExampleCode> at line 6 but never used it. While the standard component shows LLM_API_KEY setup (which does not apply here), users still need runnable instructions.
Add a bash code block after line 144:
| ``` | |
| This example does not use an LLM API key directly — the ACP server (Claude Code) handles authentication on its own. | |
| ```bash Running the Example | |
| # Ensure Claude Code CLI is authenticated first | |
| # (or set CLAUDE_API_KEY in your environment) | |
| cd software-agent-sdk | |
| uv run python examples/01_standalone_sdk/40_acp_agent_example.py |
This follows the pattern from other guides while acknowledging the different auth model.
3 participants
Summary
sdk/guides/agent-acp.mdxdocumentingACPAgentusage, configuration, metrics, and cleanupexamples/01_standalone_sdk/40_acp_agent_example.pyfrom software-agent-sdk#2133docs.jsonContext
This branch is needed by the
check-documented-examplesCI job on software-agent-sdk PR #2133, which checks out a matching docs branch (feat/acp-agent) to verify that new examples are documented.🤖 Generated with Claude Code