Skip to content

Commit d0d4c7d

Browse files
committed
Merge remote-tracking branch 'origin/main' into claude/project-thread-3jyypm
2 parents dde7b3a + d280265 commit d0d4c7d

18 files changed

Lines changed: 1087 additions & 901 deletions

‎.changeset/29341-connector-credential-literal.md‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -11,6 +11,6 @@ Metadata-protocol: none
1111
A declarative connector entry's four maps — `providerConfig.headers`, `providerConfig.defaultHeaders`, `providerConfig.transport.headers` and `providerConfig.transport.env` — are withheld whole on the metadata read path, but a literal credential in one of them is still stored in metadata. The same `credential-literal` predicate the `flow-credential-literal` advisory asks now judges those four maps and names each credential-shaped entry by its position (never its value), with the `auth.credentialRef` route:
1212

1313
- a header entry: move it to the connector's `auth: { type: 'bearer' | 'api-key' | 'basic', credentialRef }`, which the `rest` provider and an `mcp` http transport send themselves;
14-
- a stdio `transport.env` entry: a stdio MCP server takes no `auth`, so the warning says `auth.credentialRef` reaches the server only over an http transport.
14+
- a stdio `transport.env` entry: move it to the connector's `auth: { type: 'bearer' | 'api-key' | 'basic', credentialRef }` and name the variable in `providerConfig.transport.credentialEnv`; the `mcp` provider hands the resolved credential to the stdio server in that variable itself.
1515

1616
Always a `warning`; nothing is refused and every save lands. The rule runs on all three CLI commands and on the runtime publish gate for `connector` (`PUT /meta/connector/:name`), where it rides the save response's `advisories`. `os explain connector-credential-literal` carries the reasoning.
Lines changed: 18 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,18 @@
1+
---
2+
'@objectstack/connector-mcp': minor
3+
'@objectstack/lint': patch
4+
---
5+
6+
Clause-②: yes (widening)
7+
Metadata-protocol: added
8+
9+
`provider: 'mcp'`: a stdio MCP server can now take its credential from the secrets layer instead of a literal in `transport.env`. Declare the instance's `auth` with a `credentialRef` as for an http server, and name the environment variable the server reads in the new `providerConfig.transport.credentialEnv` key: the host resolves the reference (environment, Setup → Connector credentials, the host's resolver) and the provider hands the resolved secret (the `bearer` token, the `api-key` key, or the `basic` password) to the child in that variable. The value is never stored, logged, served on a metadata read or put in a message; a reference that resolves to nothing degrades the instance naming the reference and the server is not started.
10+
11+
```ts
12+
providerConfig: { transport: { kind: 'stdio', command: 'repo-mcp-server', credentialEnv: 'REPO_MCP_TOKEN' } },
13+
auth: { type: 'bearer', credentialRef: 'REPO_MCP_TOKEN' },
14+
```
15+
16+
Refused at materialization as configuration faults (a failed boot for the host's own entry): a `credentialEnv` that is not an environment-variable name, one on an `http` transport, one with no `bearer` / `api-key` / `basic` `auth`, and one naming a variable `transport.env` also sets. A stdio transport that names no `credentialEnv` is unchanged: it does not apply `auth`.
17+
18+
`@objectstack/lint`: the `connector-credential-literal` advisory's finding for a credential literal in `transport.env` now steers to this route (`auth.credentialRef` with `transport.credentialEnv`) instead of telling the author to move to an http transport.

‎.github/workflows/ci.yml‎

Lines changed: 39 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -3885,23 +3885,56 @@ jobs:
38853885
# paths relative to ui/, the vitest root, passed as positionals with no
38863886
# `--`, the only spelling ui/scripts/vitest-invocation-guard.mjs admits.
38873887
#
3888-
# objectui's full-battery runs also pass its per-file timings reporter
3889-
# and upload the result for the weekly dataset refresh. Both belong to
3890-
# that refresh workflow, which is in the governance-workflow audit
3891-
# (sub-issue 4 of the epic); until it is ported the committed dataset
3892-
# is what the sequencer reads.
3888+
# THE PER-FILE TIMINGS (objectstack#29517). The whole suite also passes
3889+
# ui/scripts/test-file-timings.mjs, the per-file timings reporter, beside
3890+
# the two reporters it would otherwise displace. It writes this shard's
3891+
# report to $OBJECTUI_TEST_TIMINGS_OUT, and the step after this one
3892+
# uploads it on the full battery's events as `test-file-timings-N`.
3893+
# .github/workflows/ui-test-timings-refresh.yml merges those reports
3894+
# weekly into ui/scripts/test-shard-timings.json, the dataset the shard
3895+
# sequencer splits on. Report-only, by ruling: the reporter never throws
3896+
# into vitest (a failure is one stderr line and no report), and nothing
3897+
# here or there grades a duration. The affected speed passes no reporter:
3898+
# a slice of the suite is not a measurement of it.
3899+
# ui/scripts/__tests__/test-file-timings.test.ts holds this step, the
3900+
# upload and the refresh's `--shards` to the names that script spells.
3901+
# The reporter is spelled from the workspace root, never `./scripts/…`
3902+
# relative to ui/: the root's workflow readers (check-self-test-wired)
3903+
# key a `./scripts/` path to the ROOT `scripts/`, where it is not.
38933904
- name: Run the ui tests (shard ${{ matrix.shard }}/16)
38943905
env:
38953906
MODE: ${{ steps.select.outputs.mode }}
3907+
OBJECTUI_TEST_TIMINGS_OUT: ${{ runner.temp }}/test-file-timings/shard-${{ matrix.shard }}.json
38963908
run: |
38973909
if [ "$MODE" = 'affected' ]; then
38983910
mapfile -t filters < "$RUNNER_TEMP/affected-tests.txt"
38993911
echo "Affected speed: ${#filters[@]} vitest filter(s), shard ${{ matrix.shard }} of 16."
39003912
pnpm --dir ui exec vitest run --shard=${{ matrix.shard }}/16 --reporter=default --reporter=github-actions "${filters[@]}"
39013913
else
3902-
pnpm --dir ui run test --shard=${{ matrix.shard }}/16 --reporter=default --reporter=github-actions
3914+
pnpm --dir ui run test --shard=${{ matrix.shard }}/16 --reporter=default --reporter=github-actions --reporter="${GITHUB_WORKSPACE}/ui/scripts/test-file-timings.mjs"
39033915
fi
39043916
3917+
# Diagnostics only, so it can never red a shard: `continue-on-error`, and
3918+
# a missing report is a warning, not an error (the refresh's selector
3919+
# then skips this run by name). Uploaded on the full battery's events
3920+
# only, the deny-list of THE TWO SPEEDS above `jobs:`; a pull request that
3921+
# failed open to the whole suite writes a report and uploads nothing.
3922+
# `overwrite: true` so a re-run shard replaces its report, as the
3923+
# shard-attestation uploads do. The refresh reads only the newest
3924+
# completed `push` / `schedule` runs of this workflow on `main`, as many
3925+
# as its selector's own cap examines, so the retention below is
3926+
# deliberately more than that window needs; it is not a measurement of it.
3927+
- name: Publish this shard's per-file test timings
3928+
if: ${{ always() && github.event_name != 'pull_request' && github.event_name != 'merge_group' }}
3929+
continue-on-error: true
3930+
uses: actions/upload-artifact@v7
3931+
with:
3932+
name: test-file-timings-${{ matrix.shard }}
3933+
path: ${{ runner.temp }}/test-file-timings/
3934+
if-no-files-found: warn
3935+
retention-days: ${{ github.event_name == 'schedule' && 7 || 2 }}
3936+
overwrite: true
3937+
39053938
test-ui-gate:
39063939
# Stable required-check name for the sharded Test UI matrix (#26460, ruling
39073940
# A, comment 6435318516), in test-gate's shape and for test-gate's reason

‎.github/workflows/fleet-write.yml‎

Lines changed: 8 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -151,11 +151,14 @@ jobs:
151151
# also reaches cancel, re-run, delete, cache and artifact endpoints; the op table has no
152152
# row for any of them, and `validate.mjs --self-test` pins that none is reachable.
153153
permission-actions: ${{ steps.validate.outputs.permission_actions }}
154-
# THREE consumers, one at a time: the advisory rows — `advisory_report` (one private report),
155-
# `advisory_collaborators_add` and `advisory_state` (one advisory's PATCH). Minted ONLY for a
156-
# stroke carrying one of them, which travels alone, the same way as `permission-actions`: the
157-
# validator writes `permission_repository_advisories=write` for it and an EMPTY value for every
158-
# other stroke. Measured 2026-11-02 with this action and a control token minted without the grant
154+
# FOUR write consumers, one at a time: the advisory rows — `advisory_report` (one private report),
155+
# `advisory_create` (one draft advisory), `advisory_collaborators_add` and `advisory_state` (one
156+
# advisory's PATCH). Minted ONLY for a stroke carrying one of them, which travels alone, the same
157+
# way as `permission-actions`: the validator writes `permission_repository_advisories=write` for it
158+
# and an EMPTY value for every other stroke. Separately, `advisory_read` (one advisory's GET, also
159+
# alone) gets a READ-ONLY mint through this same input: the validator writes
160+
# `permission_repository_advisories=read` for its stroke, never `write`. Measured 2026-11-02 with
161+
# this action and a control token minted without the grant
159162
# (run 41415522766): the narrowed mint succeeds, and the grant opens an advisory's read and PATCH.
160163
# ⚠️ EXPECTED: such a run warns "Unexpected input(s) 'permission-repository-advisories'", because
161164
# the action's `action.yml` does not declare the input; its `lib/get-permissions-from-inputs.js`

‎.github/workflows/lint.yml‎

Lines changed: 0 additions & 24 deletions
Original file line numberDiff line numberDiff line change
@@ -1495,30 +1495,6 @@ jobs:
14951495
continue-on-error: true
14961496
run: pnpm check:pm-stale-finding
14971497

1498-
# Changeset-deadline census self-test (#16850) — same family and the same
1499-
# split as every step around it. The LIVE census reads the shared board
1500-
# over the API and is REPORT-ONLY by ruling: #16850 takes option 1 (detect
1501-
# the consumption) under this lane's standing rule that a NEW scan
1502-
# population — here, issue BODIES — is censused before anything is
1503-
# enforced. So a card whose changeset window closed is a fact about the
1504-
# board, not about the PR that happens to run CI next, and failing that
1505-
# PR would punish the wrong actor.
1506-
#
1507-
# ⛔ Do not "promote" the live census into this step, and ⛔ do not give it
1508-
# a failing exit on findings. Whether the number justifies an exit
1509-
# contract at all is the question the census exists to answer, and the
1510-
# answer it produced on 2026-09-09 is in the PR that landed this file.
1511-
#
1512-
# What the self-test instruments is the MATCHING RULE and the CONTROLS —
1513-
# the two things a clean tree cannot exercise. Weakening either (counting
1514-
# the glob spelling as a target; accepting a zero without its positive
1515-
# control) leaves a production run just as green, which is the shape
1516-
# `check:self-test-wired` requires this step for.
1517-
- name: Changeset-deadline census self-test
1518-
id: changeset-deadline-census-self-test
1519-
continue-on-error: true
1520-
run: pnpm check:pm-changeset-deadline-census
1521-
15221498
# Governed-merges audit self-test (#9495) — same family, same split as
15231499
# the half-state sweeper above: the live sweep enumerates the governed
15241500
# merges since a date/ref for the PM round report — report-only by

0 commit comments

Comments
 (0)