Skip to content

feat!: canonical MCP responses and agent fidelity in 1.2.1 - #15

Merged
dertin merged 10 commits into
mainfrom
v1.2.1
Oct 4, 2026
Merged

dertin merged 10 commits into
mainfrom
v1.2.1

Conversation

@dertin

@dertin dertin commented Oct 3, 2026 •

Copy link
Copy Markdown
Owner

MCP tool responses preserve exact graph IDs, source and evidence locators, and executable continuation arguments. Version 1.2.1 delivers these facts in one schema-6 text block: JSON by default, or equivalent Markdown through --response-format markdown. Clients read content[0].text and resolve entity/relation catalog references.

Consumer-symbol associations use repository and declaration line to distinguish same-named methods. Event links retain evidence from both endpoints, and empty FTS results remain distinct from unavailable search. maxMcpToolResponseBytes bounds the complete tool result, with explicit recovery guidance for oversized responses.

Public guides and the packaged agent skill document the current contract, supported options, and version. Rust macro definitions and invocations count as syntax candidates so calls in their token bodies pass structural source validation.

Validation: the synthetic Rust macro regression failed before the fix and passes afterward; all 686 core tests and workspace Clippy pass. 50 focused hooks/plugin/MCP tests passed; all 91 local documentation links and 11 anchors resolve; Cargo package inventories exclude internal evaluation artifacts; formatting and whitespace checks pass. CI runs the complete workspace suite, MSRV, Clippy, dependency/secret checks, native platform tests, and release smoke checks.

@dertin
dertin marked this pull request as ready for review October 3, 2026 23:57
@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 3, 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 2026-10-04T00:04:21.828040Z 039327b Draft marked ready
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: 039327b9a9

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".

Comment thread crates/code-system-graph-core/src/source_graph.rs
Comment thread crates/code-system-graph-cli/src/mcp_support/agent_context.rs Outdated
@dertin dertin changed the title feat: improve agent response fidelity and add canonical delivery in 1.2.1 feat!: canonical MCP responses and agent fidelity in 1.2.1 Oct 4, 2026
@dertin
dertin merged commit d161502 into main Oct 4, 2026
10 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