[codex] Carry source thread ID in forked history

[aibrahim-oai]

Summary

• Store the immediate source thread ID alongside forked rollout history.
• Preserve that provenance through normal subagent forks, guardian snapshots, app-server resume, and internal history transformations.
• Leave imported external transcripts without a source thread ID instead of inferring one from copied metadata.

Validation

• just fmt
• Online CI

[aibrahim-oai]

This change is part of the following stack:

• [codex] Carry source thread ID in forked history #25662 ◀
  • Track app-server start in ChatGPT telemetry #24142
    • Track thread start in ChatGPT telemetry #24143
      • Track turn timing in ChatGPT telemetry #24144
  • [codex] Lock multi-agent runtime version per thread #25351
    • [codex] Select multi-agent version from model info #25427
  • [codex] Add model tool mode selector #25031
    • [codex] Add model multi-agent version selector #25032
  • [codex] Persist multi-agent version metadata #25153
    • [codex] Pin multi-agent system per thread #25168
  • Publish Linux runtime wheels with glibc-compatible tags #21812
    • [1/8] Pin Python SDK runtime dependency #21891
  • [6/8] Add high-level Python SDK approval mode #21910
  • [8/8] Add Python SDK Ruff formatting #22021
  • [2/8] Generate Python SDK types from pinned runtime #21893
    • [3/8] Run Python SDK tests in CI #21895
  • [7/8] Add Python SDK app-server integration harness #22014
  • [5/8] Rename Python SDK package to openai-codex #21905
  • [4/8] Define Python SDK public API surface #21896
  • [codex] Prepare Python SDK beta documentation and package metadata #24836
  • [codex] Add independent beta release for the Python SDK #24828
  • [1/4] Add executor HTTP request protocol #18581
    • [2/4] Implement executor HTTP request runner #18582
      • [3/4] Add executor-backed RMCP HTTP client #18583
        • [4/4] Honor Streamable HTTP MCP placement #18584
  • Use structured service tiers across core and app-server #20822
    • Drive TUI service-tier commands from model metadata #20824

_{Change managed by git-spice.}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: f8d2349c93

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".

[chatgpt-codex-connector]

Do not reuse ignored resume id as fork source

When thread/resume supplies history, the v2 contract says history takes precedence and thread_id is ignored (app-server-protocol/src/protocol/v2/thread.rs lines 321-323). Copying that parameter into ForkedHistory means any parseable placeholder or unrelated id is persisted and returned as thread.forkedFromId; clients such as the TUI then treat it as real fork lineage and immediately issue thread/read for that parent while mapping the resume response. This makes history-based resumes expose bogus parent metadata and can trigger spurious parent lookups unless the source is carried in a separate validated field or left None.

Useful? React with 👍 / 👎.