feat(docs): improve skill and documentation - #374
Conversation
Signed-off-by: Soumili Nandi <soumilin@nvidia.com>
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: ASSERTIVE Plan: Enterprise Run ID: ⛔ Files ignored due to path filters (1)
📒 Files selected for processing (1)
📜 Recent review details⏰ Context from checks skipped due to timeout. (4)
🧰 Additional context used📓 Path-based instructions (1){.github/**,ci/**,.pre-commit-config.yaml,pyproject.toml,uv.lock,mcp/pyproject.toml,mcp/uv.lock}⚙️ CodeRabbit configuration file
Files:
🔇 Additional comments (1)
WalkthroughDocumentation adds optional artifact-storage secrets and revises local NIM GPU guidance. Skill headings and cross-skill routing evaluations are updated, and the ChangesDocumentation, skill evaluation, and dependency updates
Estimated code review effort: 1 (Trivial) | ~5 minutes Possibly related PRs
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✨ 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 `@deploy/helm/README.md`:
- Around line 366-374: Update the aiq-credentials rotation procedure to preserve
or recreate all documented optional credentials alongside the four required
keys, including Modal, artifact-storage, and AWS S3 values listed in the
environment-variable table. Ensure the secret workflow remains
configuration-parity with the supported keys documented near MODAL_TOKEN_ID and
AIQ_ARTIFACT_BLOB_PROVIDER.
🪄 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: ASSERTIVE
Plan: Enterprise
Run ID: 8c000933-a262-4e63-a063-f4a6eee45033
📒 Files selected for processing (6)
deploy/helm/README.mddocs/source/examples/cli-with-local-nims.mdskills/aiq-deploy/SKILL.mdskills/aiq-deploy/evals/evals.jsonskills/aiq-research/SKILL.mdskills/aiq-research/evals/evals.json
📜 Review details
⏰ Context from checks skipped due to timeout. (2)
- GitHub Check: Run Harbor skill eval
- GitHub Check: Pytest and Coverage
🧰 Additional context used
📓 Path-based instructions (5)
docs/source/**/*
📄 CodeRabbit inference engine (AGENTS.md)
Update the docs under docs/source/ when behavior, configuration, or workflows change
Files:
docs/source/examples/cli-with-local-nims.md
{docs/**,README.md,CONTRIBUTING.md,SECURITY.md,CODE-OF-CONDUCT.md}
⚙️ CodeRabbit configuration file
{docs/**,README.md,CONTRIBUTING.md,SECURITY.md,CODE-OF-CONDUCT.md}: Review documentation for command accuracy, branch-name consistency, current CI and copy-pr-bot behavior, public
vs internal boundary clarity, stale examples, and links that no longer match the repository layout.
Files:
docs/source/examples/cli-with-local-nims.md
{deploy/**,configs/**}
⚙️ CodeRabbit configuration file
{deploy/**,configs/**}: Review deployment and config changes for secret separation, safe defaults, local-vs-production behavior, Helm and
Docker portability, and documentation parity. Flag committed credentials, environment-specific NVIDIA internals in
public defaults, and changes that make examples diverge from CI-tested paths.
Files:
deploy/helm/README.md
{skills/**,.agents/skills/**,.claude/skills/**,.github/skill-eval/**}
⚙️ CodeRabbit configuration file
{skills/**,.agents/skills/**,.claude/skills/**,.github/skill-eval/**}: Review Agent Skill and skill-eval changes for valid skill metadata, deterministic eval specs, safe handling of
credentials, and clear generated-output boundaries. Do not flag SKILL.md files for missing SPDX headers when the
entrypoint intentionally starts with YAML frontmatter.
Files:
skills/aiq-research/evals/evals.jsonskills/aiq-deploy/evals/evals.jsonskills/aiq-deploy/SKILL.mdskills/aiq-research/SKILL.md
skills/aiq-research/**/SKILL.md
📄 CodeRabbit inference engine (skills/aiq-research/SKILL.md)
skills/aiq-research/**/SKILL.md: Use this skill only for research-shaped requests; do not use it for install, deploy, start, stop, UI, CLI, Docker, Helm, or troubleshooting requests.
Run thehealthcommand before sending research requests.
Before sending a user query, state the exact AI-Q backend URL that will receive it; use non-local URLs only after the user explicitly confirms they are trusted.
Do not send credentials, cookies, bearer tokens, API keys, basic-auth credentials, or other secret values in query text orAIQ_SERVER_URL.
If no backend is reachable, ask for a backend URL or hand off toaiq-deploy; do not fabricate a research answer.
If the backend returns HTTP 401 or 403, stop and explain that this public skill does not manage authentication; do not retry until authentication is resolved.
Poll asynchronous research jobs when ajob_idis returned, and stop on failed, failure, or cancelled jobs without retrying automatically.
Present completed reports with citations and source URLs intact; do not truncate them.
Usestatus,report, orresearch_pollto resume interrupted jobs because jobs continue server-side.
For report questions answerable from an existing report, answer directly without calling the backend again; use a fresh request only when new investigation is needed.
For a redo, treat the request as a new job: state and verify the target endpoint again, then poll and present the result using the normal workflow.
Use the configuredAIQ_SERVER_URL, defaulting tohttp://localhost:8000when it is unset.
Only proceed when the Blueprint version is compatible with skill version 2.2.0: matching major version and Blueprint minor version greater than or equal to 2.2.
When materializing artifacts, use the provided artifact download or portable report commands and preserve rewritten artifact links in shareable reports.
Files:
skills/aiq-research/SKILL.md
🧠 Learnings (1)
📓 Common learnings
Learnt from: CR
Repo: NVIDIA-AI-Blueprints/aiq
Timestamp: 2026-07-21T19:31:22.628Z
Learning: When no deployment mode is specified, ask the user to choose Skill backend, CLI, UI, or Custom before starting services; do not ask this when the mode is already specified or when `aiq-research` has routed a backend request.
Learnt from: CR
Repo: NVIDIA-AI-Blueprints/aiq
Timestamp: 2026-07-21T19:31:22.628Z
Learning: After deployment, report the verified `AIQ_SERVER_URL` to `aiq-research`, and do not continue into deep research or deep-research completion validation unless the user requests or confirms it.
Learnt from: CR
Repo: NVIDIA-AI-Blueprints/aiq
Timestamp: 2026-07-21T19:31:22.628Z
Learning: When troubleshooting port conflicts, identify the process first and obtain user approval before stopping it; alternatively configure a different deployment port and revalidate health.
Learnt from: CR
Repo: NVIDIA-AI-Blueprints/aiq
Timestamp: 2026-07-21T19:31:22.628Z
Learning: Use `configs/config_web_default_llamaindex.yml` for the default Skill backend when API-enabled configuration is required by `aiq-research`.
🔇 Additional comments (5)
docs/source/examples/cli-with-local-nims.md (1)
205-206: LGTM!skills/aiq-deploy/SKILL.md (1)
27-27: LGTM!Also applies to: 62-62
skills/aiq-research/SKILL.md (1)
35-35: LGTM!Also applies to: 67-67
skills/aiq-deploy/evals/evals.json (1)
30-41: LGTM!skills/aiq-research/evals/evals.json (1)
45-56: LGTM!
Signed-off-by: Soumili Nandi <soumilin@nvidia.com>
|
/merge |
0e3cafa
into
NVIDIA-AI-Blueprints:release/2.2
2 participants
Overview
Addresses FDR feedback for the AI-Q release docs and public skills:
nvidia/nemotron-3-super-120b-a12bmodel used in the example.When to Use This SkillandWorkflowformat.aiq-deployandaiq-research.DCO sign-off for the squash commit
Signed-off-by: Soumili Nandi soumilin@nvidia.com
Validation
git commit -sor an equivalent sign-off.Validation run:
doc changes - skill validated using invoking via Agent
Where should reviewers start?
Start with:
skills/aiq-deploy/SKILL.mdskills/aiq-research/SKILL.mdskills/aiq-deploy/evals/evals.jsonskills/aiq-research/evals/evals.jsonThe docs-only deployment updates are in:
deploy/helm/README.mddocs/source/examples/cli-with-local-nims.mdRelated Issues
Summary by CodeRabbit
Documentation
Tests
Maintenance