Skip to content

[Narada Code 7] Request execution trace capture explicitly#153

Draft
xTRam1 wants to merge 3 commits into
codex/agent-maker-m6-python-project-syncfrom
codex/agent-maker-trace-opt-in
Draft

[Narada Code 7] Request execution trace capture explicitly#153
xTRam1 wants to merge 3 commits into
codex/agent-maker-m6-python-project-syncfrom
codex/agent-maker-trace-opt-in

Conversation

@xTRam1

@xTRam1 xTRam1 commented Jun 10, 2026

Copy link
Copy Markdown
Contributor

Summary

Updates CPython and Pyodide SDK remote-dispatch request construction so execution trace capture is requested only for:

  • Agent.run(trace=True)
  • an active async with trace(...) session
  • inherited NARADA_EXECUTION_TRACE_CONTEXT propagation

Normal Agent.run(...) remains backward-compatible and does not send captureExecutionTrace.

Why

SDK trace materialization is opt-in, so the SDK should not trigger frontend trace capture unless the caller wants a materializable trace proof root or is continuing a parent trace.

Validation

  • uv run ruff check packages/narada packages/narada-core packages/narada-pyodide
  • uvx ruff@latest format --check --diff packages/narada packages/narada-core packages/narada-pyodide
  • uv run pytest packages/narada/tests packages/narada-pyodide/tests/test_cloud_browser.py
  • Result: 174 passed.

Live proof

  • Signed-in DEV stack: /tmp/narada-e2e-runs/20260610T025557Z-stack-m7-trace-optin-20260609195557
  • Untraced SDK request returned no executionTraceContext: 20260610030031-7U9njSw6UDbJZobLXxafC
  • Traced SDK request returned context and materialized cleanly: /tmp/narada-workbench-runs/m7-trace-optin-traced-20260610T0300Z
  • score and verify passed on the materialized proof root.
  • Nested traced Python APA harness backstop passed: /tmp/narada-e2e-runs/20260610T031228Z-executiontrace-1c4df2245561486084520929489cab30

M7 skill dogfood note

The local narada-agent-maker skill package was created and installed, and a first Codex dogfood used it correctly for CLI discovery, scaffolding, authority/validation planning, and positive/negative local validation. I stopped the dogfood before Agent Studio sync/run because the child Codex inspected and printed a plaintext SDK-key artifact. The secret was revoked/deleted/redacted, the skill was patched to forbid reading/printing/copying plaintext key artifacts, and a clean full dogfood rerun remains the next M7 acceptance gate.

Codex sibling refs

  • caddie: branch:codex/agent-maker-trace-opt-in
  • frontend: branch:codex/agent-maker-trace-opt-in
  • architecture-docs: branch:codex/agent-maker-trace-opt-in-docs

@xTRam1 xTRam1 force-pushed the codex/agent-maker-m6-python-project-sync branch from 8855661 to 629717a Compare June 10, 2026 09:11
@xTRam1 xTRam1 force-pushed the codex/agent-maker-trace-opt-in branch from 55ada41 to 2e0b743 Compare June 10, 2026 09:12
@xTRam1 xTRam1 changed the title [Narada Code 7] Request execution trace capture explicitly [Narada Code 7] Request trace capture explicitly and package Agent Maker skill Jun 10, 2026
@xTRam1 xTRam1 force-pushed the codex/agent-maker-trace-opt-in branch from e0d5177 to 9c90cf1 Compare June 10, 2026 22:44
@xTRam1 xTRam1 changed the title [Narada Code 7] Request trace capture explicitly and package Agent Maker skill [Narada Code 7] Request execution trace capture explicitly Jun 10, 2026
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