Skip to content

docs: roadmap evidence - P520 commit-pinned Node receipt PASS (main 3c7d3d98) #868

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
kmjtechno merged 1 commit into from
Nov 9, 2026
Merged

docs: roadmap evidence - P520 commit-pinned Node receipt PASS (main 3c7d3d98) #868

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion docs/specs/KMJ-CodeBridge-Roadmap.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -574,11 +574,14 @@ evidence_log:
- fact: The OmniDesk lock-file failure is not limited to the old local commit. Upstream origin/main (22406fbe7507da5cb45bf6619367d09c917188e2) also fails cargo check --locked because Cargo.lock lacks entries Cargo.toml needs, and none of the 51 commits since d49fa215 changed Cargo.lock. A minimal regeneration adds 132 lines and removes none, after which cargo check --locked passes.
evidence: owner-run local commands on the P520 (not CodeBridge runner jobs). Upstream check exit 101; fix commit 23884fe1d08d4d0fa75d36ad925fe6928f72fa25 on a local-only branch, check --locked exit 0, no other tracked file changed.
scope: compile check only on one machine. The fix is not pushed or merged; the OmniDesk upstream CI does not use --locked, so it does not catch this. The CodeBridge cargo-standard preset does use --locked, which is why it exposed it.
- fact: The M1 runner executed a commit-pinned Node build/test on the P520 Windows desktop agent. On main 3c7d3d98531ec0e2bf304f974dcbaac2b2fa317c in isolated lane node-m1-main3, node_deps locked (npm ci) exited 0 and node_verify test exited 0 with 1959 tests, 1745 pass, 0 fail, 214 skipped.
evidence: test job 38414f19-1706-4bd2-908a-60ad1665ccf6, gate command:node_verify:test:build@node-m1-main3#3c7d3d98531ec0e2bf304f974dcbaac2b2fa317c, 138.6 s, receiptSha256 60766b6a05a7b72e13dae3a09f9ce83ac5aea6c65584d90a768e1554e169a292, logSha256 f46415db244fdc93210a3a1a3c18d378eb373ce5064c23840a52bd6e51f296d9 (log truncated to the 32 KiB tail); deps job 34bbdd97-11b0-4998-bf12-06457325ac84 exit 0. node_verify check exited 0 earlier on 38ff4ab0 (job 1f22debb-b2b3-47e5-a7c0-52afb135dd6a); it was not re-run on 3c7d3d98 (the lane is dirty after npm ci, WORKTREE_HAS_UNSAVED_WORK).
scope: one Node profile on one Windows machine. Getting here took four runs and exposed three real defects, all fixed in main, namely bare npm could not be spawned without a shell on Windows (PR 864, job 0569fdb0 exit -4058 ENOENT), and two Windows-only test assumptions (PR 865 autoconnect wrong-machine test ran on the real HomeOffice host, PR 866 and PR 867 symlink-privilege fixtures failed because the runner exports CI=true). The 214 skips include symlink assertions that NOT EXECUTED on this non-admin machine; only GitHub Windows CI runs them. The agent runtime was updated by an owner-run script (fast-forward of KMJ\codebridge-runtime to 0c2acbd4). It proves no CMake/CTest, no CineCore, no GUI/HIL and no release readiness.
not_claimed_2026_11_08:
- Autonomous coding worker (developmentWorkerReady false; no local model runtime on the checked devices). Autonomy remains milestone M8 and needs its own design, security review and cost model.
- Device-jobs flag (OFF; needs separate explicit owner approval).
- Product Manager owner QA, Workspace chat end-to-end, Google backup restore test, VPS gateway currently behind main (VPS Deployment Smoke VPS_COMMIT_MISMATCH).
- signedUpdaterVerified, production release approval, any public or paid launch.
- A PASSING commit-pinned build or any test receipt on the P520, any CMake/CTest or Node receipt there, or any CineCore receipt. The only commit-pinned receipt so far is the FAIL above; no node or cmake profile is configured on the other enrolled desktop agents, so the M1 runner is still unproven for those profiles.
- Any CMake/CTest receipt or any CineCore receipt, and any PASSING commit-pinned cargo (OmniDesk) receipt, since the OmniDesk commit-pinned receipt is still the FAIL above. The Node profile now has a passing P520 receipt (see the fact above); no cmake profile is configured on the enrolled desktop agents, so the M1 runner is still unproven for CMake/Qt/CTest.
- Mid-session watchdog respawn on the P520: the dual-agent watchdog's launcher path does not match the existing Startup shortcuts, so the launcher reports DEVICE_STARTUP_SHORTCUT_CONFLICT. Logon reconnect shortcuts do exist for the agents; the OmniDesk agent currently runs from a manually started process.
- Milestone statuses were reviewed against this evidence and left unchanged; no acceptance criterion in M0 to M8 was newly proven.