Skip to content

Load NyxID recommended skill refs - #3665

Merged
louis4li merged 15 commits into
feature/integratefrom
feat/2026-09-18_nyxid-recommended-skill-read
Sep 30, 2026
Merged

louis4li merged 15 commits into
feature/integratefrom
feat/2026-09-18_nyxid-recommended-skill-read

Conversation

@louis4li

Copy link
Copy Markdown
Contributor

Issue

Closes #3661

Scope

  • Add typed NyxID connected-service recommended skill refs to the inventory contract and parser.
  • Expose a Channel-only nyxid_load_recommended_skill reader that revalidates sender inventory, loads exact Ornn refs, and verifies manifest digests.
  • Update Channel prompt guidance and focused tests for exact recommended-skill loading.
  • Refresh the NyxID assistant conformance Aevatar source pin required by the architecture guard.

Verification

  • dotnet build agents/Aevatar.GAgents.NyxidChat/Aevatar.GAgents.NyxidChat.csproj --nologo
  • dotnet test test/Aevatar.AI.Tests/Aevatar.AI.Tests.csproj --nologo --filter "NyxIdServiceInstanceClientTests|NyxIdConnectedServiceToolSourceTests"
  • dotnet test test/Aevatar.GAgents.ChannelRuntime.Tests/Aevatar.GAgents.ChannelRuntime.Tests.csproj --nologo --filter ChannelNyxIdConnectedServiceInventoryToolSourceTests
  • bash tools/ci/test_stability_guards.sh
  • bash tools/ci/architecture_guards.sh
  • Live NyxID deployment endpoint check: not run locally; requires deployed NyxID inventory exposing non-empty recommended_skill_refs.

Notes

  • The loader does not fall back to display names, latest versions, hidden catalog backfill, or raw proxy calls.
  • The recommended skill document is loaded as guidance only; service execution still goes through admitted connected-service tools and frozen schemas.

Generated with Claude Code.

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Sep 18, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-09-18T04:48:35.301017Z a63449f PR opened
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@chatgpt-codex-connector chatgpt-codex-connector Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: a63449f88d

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

@louis4li
louis4li force-pushed the feat/2026-09-18_nyxid-recommended-skill-read branch from c66139e to 4e3ffeb Compare September 30, 2026 09:26
louis4li and others added 15 commits September 30, 2026 17:27
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace dynamic connected-service endpoint exposure with the recommended-skill inventory, load, and fixed invocation path, and generate missing Ornn skills from NyxID operation contracts.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Keep inventory read-only, require exact recommended refs for document-guided requests, and expose explicit provisioning for missing refs while preserving NyxID as the source of truth.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Keep recommended Ornn skill refs available when channel inventory responses omit them, so reply generation can continue using committed recommended skill references after the latest feature/integrate merge.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Keep Ornn publish failures typed after the rebase and align channel runtime tests with the current registration-key credential and inventory-source behavior.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@louis4li
louis4li force-pushed the feat/2026-09-18_nyxid-recommended-skill-read branch from 4e3ffeb to 067a5bc Compare September 30, 2026 09:36
@louis4li
louis4li merged commit aaf4b22 into feature/integrate Sep 30, 2026
12 of 15 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant