[Narada Code 7] Request execution trace capture explicitly#153
Draft
xTRam1 wants to merge 3 commits into
Draft
Conversation
8855661 to
629717a
Compare
55ada41 to
2e0b743
Compare
e0d5177 to
9c90cf1
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Updates CPython and Pyodide SDK remote-dispatch request construction so execution trace capture is requested only for:
Agent.run(trace=True)async with trace(...)sessionNARADA_EXECUTION_TRACE_CONTEXTpropagationNormal
Agent.run(...)remains backward-compatible and does not sendcaptureExecutionTrace.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-pyodideuvx ruff@latest format --check --diff packages/narada packages/narada-core packages/narada-pyodideuv run pytest packages/narada/tests packages/narada-pyodide/tests/test_cloud_browser.pyLive proof
/tmp/narada-e2e-runs/20260610T025557Z-stack-m7-trace-optin-20260609195557executionTraceContext:20260610030031-7U9njSw6UDbJZobLXxafC/tmp/narada-workbench-runs/m7-trace-optin-traced-20260610T0300Zscoreandverifypassed on the materialized proof root./tmp/narada-e2e-runs/20260610T031228Z-executiontrace-1c4df2245561486084520929489cab30M7 skill dogfood note
The local
narada-agent-makerskill 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
branch:codex/agent-maker-trace-opt-inbranch:codex/agent-maker-trace-opt-inbranch:codex/agent-maker-trace-opt-in-docs