[M3gA-Mind]

Summary

Run directly against a staged TinySecurity, the agent harness E2E (tests/agent_harness_e2e.rs) hung for 120 s at sign-in in every test. This PR makes the harness answer the credential store's keyring prompts itself, so its tests get past sign-in.

Problem

Since #8536, signing in stores the session token in the OS keyring, and TinySecurity asks the user to approve that write. Tests answer it through the test-only keyring responder (test-keyring-consent, compiled into the CLI test targets). That responder acts only when OPENHUMAN_TEST_KEYRING_CONSENT is set, and only scripts/test-rust-e2e.sh and scripts/ci/rust-coverage.sh set it. Run directly with cargo test --test agent_harness_e2e, with security-module and a staged TinySecurity, the boot's auth_store_session waits on a prompt nobody answers. Each test fails after its 120 s timeout (error sending request, agent_harness_e2e.rs:636) (finding TS-E2E-018).

Fix

boot_stack_with_options sets OPENHUMAN_TEST_KEYRING_CONSENT=approve for its own test process, unless the run already set a value. The responder still answers only the credential store's prompts, and it still exists only in test builds.

Scope

This removes the sign-in hang only. Past sign-in, the harness hits TinySecurity refusals it doesn't account for yet: the prompt screen has no model under the scripted mock, the temporary action directory is a protected location, and a LocalOnly refusal of the scripted model. So against a staged TinySecurity most of its tests still fail, approval_gate_approve_flow among them. Those gaps are tracked as TS-E2E-034. The official runner (scripts/test-rust-e2e.sh) doesn't stage TinySecurity and already sets the variable, so its runs are unchanged.

Tests

• Run directly with security-module, a staged TinySecurity and OPENHUMAN_TEST_KEYRING_CONSENT unset, no test in agent_harness_e2e times out at sign-in.
• Red control: on main the same run fails at sign-in (:636, error sending request after 120 s).

Validation

Cuttlefish job CF-W4-h18 on b0f4d8bd41 (base 6043171902), with a staged TinySecurity, security-module, and OPENHUMAN_TEST_KEYRING_CONSENT unset:

Check	Result
fmt	pass
clippy (cli --tests, -D warnings)	red only on main's type_complexity error at core/observability_reporting.rs:535, which stops the build before the CLI tests compile; it's in every clippy run of main 6043171902 this session
whole agent_harness_e2e, head	0 sign-in timeouts, 0 error sending request. 2 passed, 25 failed and 1 ignored, every failure after sign-in from the TinySecurity refusals above (TS-E2E-034)
red: approval_gate_approve_flow on main	fails at sign-in: :636, error sending request, after 121 s

Related: TS-E2E-018, TS-E2E-034.

Submission Checklist

• Tests added or updated (happy path + at least one failure / edge case) per Testing Strategy: the harness change is proven by agent_harness_e2e itself; the red control (main's harness) times out at sign-in (:636, "error sending request" after 121 s).
• N/A: diff coverage. Test-only change.
• N/A: no feature row added, removed or renamed in docs/TEST-COVERAGE-MATRIX.md.
• N/A: no matrix feature IDs affected; the review finding is listed under Related.
• No new external network dependencies introduced.
• N/A: no release-cut surface changes (docs/RELEASE-MANUAL-SMOKE.md).
• N/A: no GitHub issue tracks this. It is TinySecurity E2E review finding TS-E2E-018 (follow-up gaps: TS-E2E-034).

---

AI Authored PR Metadata (required for Codex/Linear PRs)

Linear Issue

• Key: N/A
• URL: N/A

Commit & Branch

• Branch: test/agent-harness-keyring-consent
• Commit SHA: b0f4d8bd41a9dfc08251f5e2aa4283f1db57f780

Validation Run

• N/A: pnpm --filter openhuman-app format:check, no frontend change
• N/A: pnpm typecheck, no TypeScript change
• Focused tests: on cuttlefish (CF-W4-h18)
• Rust fmt/check (if changed): on cuttlefish (CF-W4-h18)
• N/A: Tauri fmt/check, no shell change

Validation Blocked

• command: N/A
• error: N/A
• impact: N/A

Behavior Changes

• Intended behavior change: none in product code; the test harness answers keyring prompts.
• User-visible effect: none.

Parity Contract

• Legacy behavior preserved: all product paths; test-only.
• Guard/fallback/dispatch parity checks: N/A

Duplicate / Superseded PR Handling

• Duplicate PR(s): none
• Canonical PR: this
• Resolution (closed/superseded/updated): N/A

[coderabbitai]

Warning

Review limit reached

• Run on-demand review

_{This review includes 1 billable file and costs up to $0.25.}

Or wait 48 minutes for your next included review.

Check out review usage here.

View limit details

Limit details: You’ve used all 10 included reviews currently available.

Learn how review limits work.

Review configuration:

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: c9a7d906-3f8e-4689-8379-6c5de8b3159c

Commits

Reviewing files that changed from the base of the PR and between 6043171 and b0f4d8b.

Files selected for processing: 1

• tests/agent_harness_e2e.rs

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

[chatgpt-codex-connector]

Codex Review Summary

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

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-11-06T06:09:25.711964Z	b0f4d8b	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.

[tinysweeper]

Tiny Sweeper review

Reviewed head: b0f4d8bd41a9dfc08251f5e2aa4283f1db57f780

Verdict: Ready for maintainer review · Priority: none
Coverage: Reviewed 1 of 1 files.
Pull request purpose (title): test(agent): answer…

Diagnostics and supporting details

Full summary

Pull request purpose (title): test(agent): answer the credential store's keyring prompts in the harness. The test-only responder still answers only the credential store's prompts and remains absent from product builds, so product code and user-visible behavior are unchanged. Latest-push delta: unavailable — no trusted base-to-head diff comparison was available, so no push-delta behavior is asserted.

Review snapshot

Change surface	Files	Review signal	Count
Production	0	Active findings	0
Tests	1	Noted findings	0
Documentation	0	Resolved findings	0
Configuration	0	Pending checks/questions	0

Completeness: Complete
Test assessment: No supported feature-to-test mapping was available; this does not mean tests are absent or passed.

What changed

No product code or user-visible behavior changes. Latest-push delta: unavailable — the prior reviewed head exists, but there was no trusted base-to-head diff comparison, and cached prompt differences are not a proven push delta.

Features

None identified with supported citations.

Tests

No supported feature-to-test mapping was produced. Test execution is not inferred.

Findings

No active actionable findings.

Before merge

None.

Agent review details

critique

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Lane summary: The change sets OPENHUMAN_TEST_KEYRING_CONSENT to "approve" before the stack boots, matching what scripts/test-rust-e2e.sh already exports and only when the variable is unset. I could not see the agent-harness binary's process model or the test-responder build gating, but the same defaulting is already done by the repository's own E2E scripts, and the change only affects this test binary. No defect identified. _The code index is behind this pull request (indexed at `3aa1ac215243`), so retrieved context may be out of date._

security

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Positive: Security lane answered with no finding: the change is test-build plumbing backed by the existing test_responder, does not affect product builds, and introduces no attacker-controlled path. It flagged an unconfirmed pre-existing caveat that std::env::set_var on a non-main thread is unsound on newer Rust/Cargo, and this adds a second such call site in the same binary; the lane did not confirm the toolchain and treated it as a caveat rather than a finding.
• Lane summary: This change sets OPENHUMAN_TEST_KEYRING_CONSENT to "approve" in the test harness boot only when unset, which is test-build plumbing backed by the existing test_responder (which answers only when the variable says "approve" and only for the credential-store caller). It does not affect product builds and introduces no attacker-controlled path. A minor pre-existing caveat is that where an inline env assertion already ran, std::env::set_var on a non-main thread is unsound on newer Rust/Cargo, and this new call adds a second such call site in the same binary; I did not confirm the toolchain from the tree, so I flag it as a caveat rather than a finding. _The code index is behind this pull request (indexed at `3aa1ac215243`), so retrieved context may be out of date._

tests

• Conclusion: Neutral
• Scope reviewed: all assigned evidence
• Lane summary: No behavioural change: nothing outside documentation, configuration and tests.

commits

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Positive: Commits lane answered: no sensitive content in committed files, with a stale-index caveat.
• Lane summary: No sensitive content in committed files. _The code index is behind this pull request (indexed at `3aa1ac215243`), so retrieved context may be out of date._

description

• Conclusion: Success
• Scope reviewed: all assigned evidence
• Lane summary: The diff sets `OPENHUMAN_TEST_KEYRING_CONSENT=approve` in the harness's own boot helper when the variable is unset, which matches the described fix exactly and does nothing in product code; the description accurately covers the change, the test-only scope, and the remaining TS-E2E-034 gaps. No defects found. _The code index is behind this pull request (indexed at `3aa1ac215243`), so retrieved context may be out of date._

Evidence and run details

• Models: ladder/vectors-oai3, deepseek-v4.1-flash
• Spend: $0.000551
• Tokens: 54421 input · 4123 output · 11264 cached · 73 embedding
• Updated: 2026-11-06T06:15:38Z

Head	State	Pass summary
b0f4d8bd41a9	ready for maintainer review	0 active finding(s), 0 resolved finding(s) (at 2026-11-06T06:15:38Z)

[tinysweeper]

tinysweeper found nothing blocking. Approving.

Diagnostics

Total spend: $0.0006