Skip to content

[None][docs] Add README for custom Claude Code skills and agents#12920

Merged
kaiyux merged 4 commits into
NVIDIA:mainfrom
kaiyux:user/kaiyu/skills_readme
Apr 10, 2026
Merged

[None][docs] Add README for custom Claude Code skills and agents#12920
kaiyux merged 4 commits into
NVIDIA:mainfrom
kaiyux:user/kaiyu/skills_readme

Conversation

@kaiyux
Copy link
Copy Markdown
Member

@kaiyux kaiyux commented Apr 10, 2026 •
edited
Loading

Summary

  • Add .claude/README.md documenting the custom Claude Code skills and sub-agents configured for the TensorRT-LLM repo
  • Covers background on skills/agents, how they are discovered and loaded, and usage via manual invoke or automatic dispatch

Test plan

  • Verify the README renders correctly on GitHub
  • Confirm skills/agents are discoverable as described (/skills, /agents in Claude Code)

🤖 Generated with Claude Code

Summary by CodeRabbit

  • Documentation
    • Added comprehensive documentation detailing available AI-powered code skills and sub-agents, including their invocation methods, verification procedures, and usage patterns.

@coderabbitai
Copy link
Copy Markdown
Contributor

coderabbitai Bot commented Apr 10, 2026 •
edited
Loading

📝 Walkthrough

Walkthrough

A new documentation file .claude/README.md is added to the TensorRT-LLM repository, describing Claude Code skills and sub-agents conventions, including how to discover, verify, and invoke these components.

Changes

Cohort / File(s) Summary
Documentation
.claude/README.md		 New README documenting Claude Code skills and sub-agents used in the repository, explaining the distinction between skills and agents, discovery conventions, verification methods via /skills and /agents commands, and invocation modes (manual via /<skill-name>, delegation via @<agent-name>, and automatic dispatch).

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~2 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name Status Explanation Resolution
Description check ⚠️ Warning The PR description lacks required sections from the template: Description and Test Coverage sections are missing or incomplete, and the PR Checklist is not addressed. Add a formal Description section explaining the changes and reasoning, provide a Test Coverage section listing relevant tests, and complete the PR Checklist by confirming each item or marking as N/A.
✅ Passed checks (2 passed)
Check name Status Explanation
Title check ✅ Passed The title clearly and concisely describes the main change: adding a README document for Claude Code skills and agents in the .claude directory.
Docstring Coverage ✅ Passed No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✏️ Tip: You can configure your own custom pre-merge checks in the settings.

✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests

Comment @coderabbitai help to get the list of available commands and usage tips.

coderabbitai[bot]
coderabbitai Bot reviewed Apr 10, 2026
View reviewed changes
Copy link
Copy Markdown
Contributor

@coderabbitai coderabbitai Bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents 
Verify each finding against the current code and only fix it if needed.

Inline comments:
In @.claude/README.md:
- Around line 22-23: Pick one model and make the README consistent: either (A)
allow direct agent invocation or (B) enforce automatic delegation. If you choose
direct invocation, replace the sentence "never invoke agents directly" with
wording that agents may be invoked using the @<agent-name> syntax and update the
surrounding explanation to show the `@<agent-name>` example and when delegation
still occurs; if you choose delegation, remove or rework the `@<agent-name>`
example and clarify that Claude delegates to specialists and operators should
not call agents directly. Update the text instances that contain the exact
phrase "never invoke agents directly" and the `@<agent-name>` example so both
agree on the chosen model.
🪄 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: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 15df5f2e-a819-4129-be6c-e310dd4c832e

📥 Commits

Reviewing files that changed from the base of the PR and between d71e880 and 6e934f0.

📒 Files selected for processing (1)
  • .claude/README.md

Comment thread .claude/README.md Outdated
@kaiyux
[None][docs] Add README for custom Claude Code skills and agents
1edf0ca 
Document the custom skills and sub-agents configured for the TensorRT-LLM
repo, including background on how skills/agents work, how they are loaded,
and how to use them via manual invoke or automatic dispatch.

Signed-off-by: Kaiyu Xie <[email protected]>

Minor fix

Signed-off-by: Kaiyu Xie <[email protected]>

Minor fix

Signed-off-by: Kaiyu Xie <[email protected]>

clean up

Signed-off-by: Kaiyu Xie <[email protected]>
@kaiyux kaiyux force-pushed the user/kaiyu/skills_readme branch from e9bd6e2 to 1edf0ca Compare April 10, 2026 06:47
kaiyux and others added 2 commits April 10, 2026 00:05
@kaiyux @claude
[None][docs] Add naming convention for skill and agent domain prefixes
6da38ee 
Co-Authored-By: Claude Opus 4.6 (1M context) <[email protected]>
Signed-off-by: Kaiyu Xie <[email protected]>
@kaiyux
[None][docs] Consolidate naming convention section in skills README
6aff0f1 
Merge boundary notes into the prefix table definitions and collapse
the numbered naming rules into a concise guidelines list.

Signed-off-by: Kaiyu Xie <[email protected]>
QiJune
QiJune approved these changes Apr 10, 2026
View reviewed changes
Copy link
Copy Markdown
Collaborator

@QiJune QiJune left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@kaiyux
Minor fixes
5fae229 
Signed-off-by: Kaiyu Xie <[email protected]>
@kaiyux
Copy link
Copy Markdown
Member Author

kaiyux commented Apr 10, 2026

/bot skip --comment "doc changes"

@kaiyux kaiyux enabled auto-merge (squash) April 10, 2026 07:28
@kaiyux
Copy link
Copy Markdown
Member Author

kaiyux commented Apr 10, 2026

/bot skip --comment "doc changes"

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 10, 2026

PR_Github #42688 [ skip ] triggered by Bot. Commit: 5fae229 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 10, 2026

PR_Github #42691 [ skip ] triggered by Bot. Commit: 5fae229 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 10, 2026

PR_Github #42688 [ skip ] completed with state ABORTED. Commit: 5fae229

Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 10, 2026

PR_Github #42691 [ skip ] completed with state SUCCESS. Commit: 5fae229
Skipping testing for commit 5fae229

Link to invocation

@kaiyux kaiyux merged commit 4811704 into NVIDIA:main Apr 10, 2026
7 of 9 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@coderabbitai coderabbitai[bot] coderabbitai[bot] left review comments

@QiJune QiJune QiJune approved these changes

Assignees

@kaiyux kaiyux

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@kaiyux @tensorrt-cicd @QiJune