FEATURE : Deployment of Agent using Kagenti#108
Conversation
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughThis PR enables Kagenti-managed deployment of A2A agents by adding required Helm labels, introducing AgentRuntime resources that reference the deployments, tightening the a2a-sdk dependency range, and providing detailed integration documentation with an automated verification script and troubleshooting steps. ChangesKagenti Integration for Agent Deployment
🎯 3 (Moderate) | ⏱️ ~20 minutes 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@agents/a2a/langgraph_crewai_agent/AgentRuntime.yaml`:
- Line 5: The AgentRuntime.yaml hardcodes namespace: default which can break
deployments when users deploy to a different namespace; update AgentRuntime.yaml
so the namespace is not hardcoded by either adding a clear comment above the
namespace field (# Update namespace to match your deployment namespace),
parameterizing it (e.g., support a template or Kustomize var for namespace), or
remove the field to rely on the current kubectl context; also add a short note
in README.md near the Kagenti/AgentRuntime section instructing users to set the
namespace to match their `make deploy` target (reference AgentRuntime.yaml, the
namespace field, and Kagenti/AgentCards in the docs).
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Enterprise
Run ID: 3b10a16c-25b4-405b-b71c-a7ebee867760
📒 Files selected for processing (6)
.claude/skills/kagenti-deploy/SKILL.mdagents/a2a/langgraph_crewai_agent/AgentRuntime.yamlagents/a2a/langgraph_crewai_agent/README.mdagents/a2a/langgraph_crewai_agent/pyproject.tomlcharts/a2a-langgraph-crewai/templates/deployment-crew.yamlcharts/a2a-langgraph-crewai/templates/deployment-langgraph.yaml
andrewdonheiser
left a comment
andrewdonheiser
left a comment
There was a problem hiding this comment.
PR Review Summary
Thanks for adding kagenti deployment support! The SKILL.md is comprehensive and the overall approach is solid. There are a few issues to address before this can merge — most notably a missing label that will prevent kagenti discovery from working.
Findings: 1 Critical, 5 Important, 6 Suggestions
See inline comments for details.
Strengths
- Comprehensive end-to-end workflow with logical step progression
- Good enforcement gate pattern (Claude stops and asks if labels are missing)
- Correct multi-deployment AgentRuntime handling for langgraph + crew
- Accurate AgentRuntime/AgentCard relationship documentation
- Defensive
a2a-sdkversion pin aligns with repo patterns - Properly scoped — kagenti is optional, existing behavior unchanged
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@agents/a2a/langgraph_crewai_agent/README.md`:
- Around line 210-225: The Kagenti README section has Markdown lint errors
(MD032/MD031) due to missing blank lines around lists and fenced code blocks;
fix it by adding a blank line before and after each fenced block and before the
bulleted lists under the "Prerequisites:", "Apply AgentRuntime:" and "Verify:"
headings so the code fences containing `oc apply -f AgentRuntime.yaml` and `oc
get agentcard -n <your-namespace>` are each separated by an empty line from
surrounding paragraphs/lists and the "Apply AgentRuntime:" ordered steps; ensure
the three headers ("Prerequisites:", "Apply AgentRuntime:", "Verify:") are
followed by proper blank lines so markdownlint passes.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Enterprise
Run ID: e44e0392-c471-4349-b5d2-5ba3a249e677
📒 Files selected for processing (4)
.claude/skills/kagenti-deploy/SKILL.mdagents/a2a/langgraph_crewai_agent/AgentRuntime.yamlagents/a2a/langgraph_crewai_agent/README.mdagents/a2a/langgraph_crewai_agent/pyproject.toml
✅ Files skipped from review due to trivial changes (1)
- .claude/skills/kagenti-deploy/SKILL.md
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
agents/a2a/langgraph_crewai_agent/README.md (1)
231-233:⚠️ Potential issue | 🟠 Major | ⚡ Quick winRemove the extra blank line to unblock lint.
Line 232 has two consecutive blank lines, which triggers MD012 and keeps CI red.
Suggested patch
AgentCards for both LangGraph and CrewAI will appear in Kagenti UI. - #### Verify deploymentAs per coding guidelines, "Focus on major issues impacting performance, readability, maintainability and security. Avoid nitpicks and avoid verbosity."
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@agents/a2a/langgraph_crewai_agent/README.md` around lines 231 - 233, Remove the extra consecutive blank line after the "#### Verify deployment" heading to satisfy MD012 linting; open the README section containing the "#### Verify deployment" header and delete the redundant blank line so there is only a single blank line (or none, as your style requires) directly following that heading.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Outside diff comments:
In `@agents/a2a/langgraph_crewai_agent/README.md`:
- Around line 231-233: Remove the extra consecutive blank line after the "####
Verify deployment" heading to satisfy MD012 linting; open the README section
containing the "#### Verify deployment" header and delete the redundant blank
line so there is only a single blank line (or none, as your style requires)
directly following that heading.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Enterprise
Run ID: 0af5958a-9c61-4a02-b052-1980474c7bf9
📒 Files selected for processing (1)
agents/a2a/langgraph_crewai_agent/README.md
andrewdonheiser
left a comment
andrewdonheiser
left a comment
There was a problem hiding this comment.
Re-review: 7 of 12 items from the first review have been addressed — resolved those threads. One new issue found in the fix-up commit (see inline).
andrewdonheiser
left a comment
andrewdonheiser
left a comment
There was a problem hiding this comment.
Re-review after fix-up commits
8 of 13 original findings have been addressed — nice work on the turnaround. 5 items remain (1 critical, 2 important, 2 suggestions). See inline comments.
Resolved
sed -imacOS portability — SKILL.md now uses inlinecat <<EOF- Emojis in verification script — now uses
[PASS]/[FAIL]/[WARN] - Heading level inconsistency — changed to
#### - README lacking prerequisites/verification — added
template.envclarification — added to Step 2make deploy NAMESPACE=— now usesoc project- Version specifier ordering — now
>=0.3.25,<1.0.0 - Trailing comma in version specifier — fixed
shricharan-ks
force-pushed
the
feature/RHAIENG-5034
branch
from
f274c78 to
94dedde
Compare
Address PR comments Fix markdown MD032/MD031 linting issues Fix markdown MD032/MD031 linting issues Fix version in pyproject.toml Fix version in pyproject.toml Fix PR Comments removed trailing whitespace
shricharan-ks
force-pushed
the
feature/RHAIENG-5034
branch
from
94dedde to
92442bf
Compare
andrewdonheiser
left a comment
andrewdonheiser
left a comment
There was a problem hiding this comment.
All findings from both review rounds have been addressed. Nice work on the turnaround.
Verified resolved:
- Version specifier fixed (trailing comma removed)
namespacefield removed from AgentRuntime.yaml (uses oc context)- SKILL.md correctly documents controller-managed vs chart-managed labels
- Verification script extracted, complete flow removed
- All other items (sed portability, emojis, heading levels, README prerequisites, etc.)
CI is green. LGTM.
Thanks @andrewdonheiser your comments helped in maintaining the quality standards. Much appreciated 🎉 |
2 participants
Description
This feature setups the documentation and the claude skills for the deployment of an agent and using kagenti for agent visibility.
Jira Ticket
RHAIENG-5034
Testing
make testpasses (run from the affected agent directory)Checklist
.envor secret files are included in this PR