Commit Graph
4997 Commits
Author SHA1 Message Date
DottaandPaperclip 01ad83057b Merge verified native controls into Hermes qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 07:08:50 -05:00
DottaandPaperclip f01a5f02f8 fix(hermes): bound question responses and await native steering identity
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 07:08:06 -05:00
Dotta 8196aa179c Merge branch 'codex/hermes-routines' into codex/hermes-native-runner
* codex/hermes-routines:
  test(ui): handle delayed announcements for both sidebar links
2026-10-09 07:05:16 -05:00
DottaandPaperclip 547cdf59c3 test(ui): handle delayed announcements for both sidebar links
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 07:05:16 -05:00
DottaandPaperclip 84aa0fad11 Merge verified native assistant channel preservation
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 06:35:01 -05:00
DottaandPaperclip 7be8858576 Merge verified native assistant channel preservation
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-routines:
  fix(runner): preserve validated native assistant delta channels
2026-10-09 06:34:17 -05:00
DottaandPaperclip 58cb562866 fix(runner): preserve validated native assistant delta channels
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 06:33:59 -05:00
DottaandPaperclip dc53d4d6b7 fix(hermes): retain reported charges from interrupted requests
Keep authenticated verified OpenRouter usage frames in the known subtotal even when transport fails, without certifying complete cost or token totals. Cover the real pinned SDK interruption plus a successful retry and update both reviewed runtime closure pins.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 06:20:15 -05:00
DottaandPaperclip 6b554641a1 Merge shared browser fix into Hermes qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-native-runner:
  test(ui): dismiss late announcements during company navigation
2026-10-09 06:20:14 -05:00
DottaandPaperclip 8cb990b266 Merge verified announcement timing fix into native Hermes runner
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-routines:
  test(ui): dismiss late announcements during company navigation
2026-10-09 06:20:10 -05:00
DottaandPaperclip 7d0a36d393 test(ui): dismiss late announcements during company navigation
Retry the normal Agents link within a bounded deadline and use the ordinary dismissal control if the asynchronously loaded announcement covers it. Preserve the company-isolation and history assertions.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 06:20:07 -05:00
DottaandPaperclip 7b6be369f4 Merge bounded selection into Hermes qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-native-runner:
  fix(ui): bound manual agent selection to the switch transition
2026-10-09 05:46:30 -05:00
DottaandPaperclip ab98588317 Merge bounded selection and verified cold CLI checks
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-routines:
  fix(ui): bound manual agent selection to the switch transition
2026-10-09 05:46:27 -05:00
DottaandPaperclip 30d57b2971 fix(ui): bound manual agent selection to the switch transition
Reproduce the page/sidebar mismatch after a real manual switch and browser history navigation. Dismiss covering announcements through the normal UI. Preserve dense ZIP bytes with a CRC table and bound cold CLI health inside the existing setup budget.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 05:46:23 -05:00
DottaandPaperclip c0abc7caf6 Merge verified shared fixes into Hermes qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-native-runner:
  fix(ui): scope canonical agents to their authorized company
  test(server): drain runtime writes and load routes in setup
2026-10-09 05:20:41 -05:00
DottaandPaperclip 295488bf19 Merge verified company isolation and runtime teardown fixes
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-routines:
  fix(ui): scope canonical agents to their authorized company
  test(server): drain runtime writes and load routes in setup
2026-10-09 05:20:38 -05:00
DottaandPaperclip a7843be973 fix(ui): scope canonical agents to their authorized company
Keep alias caches and redirects on the agent company. Defer remembered UUID paths until canonical resolution and verify organization switching with a delayed public alias response.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 05:20:35 -05:00
DottaandPaperclip bb8b18d19e test(server): drain runtime writes and load routes in setup
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 05:20:33 -05:00
Dotta 85367ea3dc Merge branch 'codex/hermes-native-runner' into codex/hermes-qualification
* codex/hermes-native-runner:
  fix(ui): preserve exact retry feedback across agent aliases
2026-10-09 04:10:28 -05:00
Dotta cf595b8aa3 Merge branch 'codex/hermes-routines' into codex/hermes-native-runner
* codex/hermes-routines:
  fix(ui): preserve exact retry feedback across agent aliases
2026-10-09 04:10:28 -05:00
Dotta fb725383b9 fix(ui): preserve exact retry feedback across agent aliases 2026-10-09 04:10:28 -05:00
Dotta 4f40b5b666 Merge branch 'codex/hermes-native-runner' into codex/hermes-qualification
* codex/hermes-native-runner:
  test(db): set PostgreSQL deadlines before suite declarations
2026-10-09 02:53:57 -05:00
Dotta a8805d61af Merge branch 'codex/hermes-routines' into codex/hermes-native-runner
* codex/hermes-routines:
  test(db): set PostgreSQL deadlines before suite declarations
2026-10-09 02:53:53 -05:00
Dotta 7a03294664 test(db): set PostgreSQL deadlines before suite declarations 2026-10-09 02:53:47 -05:00
Dotta 7de8063a04 Merge branch 'codex/hermes-native-runner' into codex/hermes-qualification
* codex/hermes-native-runner:
  test: separate cold route setup and bound backup integration cases
2026-10-09 02:46:39 -05:00
Dotta 43697be47c Merge branch 'codex/hermes-routines' into codex/hermes-native-runner
* codex/hermes-routines:
  test: separate cold route setup and bound backup integration cases
2026-10-09 02:46:31 -05:00
Dotta ffa25aa9a3 test: separate cold route setup and bound backup integration cases 2026-10-09 02:46:28 -05:00
Dotta c87bb7017d Merge branch 'codex/hermes-native-runner' into codex/hermes-qualification
* codex/hermes-native-runner:
  fix(runner): carry steering identity into Hermes parent
  test(server): load feedback routes within suite setup
2026-10-09 02:18:42 -05:00
DottaandPaperclip 3fce49c5b9 fix(runner): carry steering identity into Hermes parent
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 02:18:38 -05:00
Dotta 660770f38c Merge branch 'codex/hermes-routines' into codex/hermes-native-runner
* codex/hermes-routines:
  test(server): load feedback routes within suite setup
2026-10-09 02:16:37 -05:00
DottaandPaperclip 74e54b713b test(server): load feedback routes within suite setup
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 02:16:33 -05:00
DottaandPaperclip 930860cd74 test(cursor): retain baseline setup checks in core qualification
Move only the additional Hermes setup assertions to the stacked follow-up. Keep all existing Cursor setup tests in place.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 01:31:25 -05:00
DottaandPaperclip 0afd50696d fix(hermes): bind cloud image identity to public setup helpers
Hash all three Hermes setup inputs. Keep the PR below the review limit by moving the unchanged public setup and account-cache test files to the stacked qualification follow-up.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 01:30:25 -05:00
DottaandPaperclip ef0f626781 Merge Hermes stack worker cleanup
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-native-runner:
  fix(test): remove immutable runtime contexts at worker exit
2026-10-09 01:24:43 -05:00
DottaandPaperclip a6672a81be Merge Hermes foundation worker cleanup
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-routines:
  fix(test): remove immutable runtime contexts at worker exit
2026-10-09 01:24:42 -05:00
DottaandPaperclip 0effdedd07 fix(test): remove immutable runtime contexts at worker exit
Restore directory write access inside the private worker home before removal, without following external directory symlinks.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-09 01:24:30 -05:00
Dotta 7ab7747fe3 Merge synchronized native parent Hermes pins
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-native-runner:
  fix(hermes): synchronize native parent distribution pins

# Conflicts:
#	packages/paperclip-runner/src/drivers/acpx/hermes-installation.ts
2026-10-08 23:21:06 -05:00
DottaandPaperclip 48bea8a9a8 fix(hermes): synchronize native parent distribution pins
Refresh both source-owned native parent pins for the sandbox payload change. Retain the prior reviewed interpreter and dependency entries; only tool_process.py changes.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 23:20:11 -05:00
DottaandPaperclip 6136eb1397 fix(hermes): pin the Linux device sandbox closure
Derive platform pins from previously verified manifests with only tool_process.py changed. Require fresh cloud materialization and retain the prior concurrent-write failure.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 23:07:47 -05:00
DottaandPaperclip a9e033c4e7 Merge the native Linux device sandbox fix into Hermes qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing>

* codex/hermes-native-runner:
  fix(hermes): mount usable Linux sandbox devices
2026-10-08 23:06:31 -05:00
DottaandPaperclip 4e050e9650 fix(hermes): mount usable Linux sandbox devices
Provide the minimal bubblewrap device mount for native shell redirects and atomic writes, probe it before credential staging, and cover concurrent writes plus protected paths.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 23:06:10 -05:00
DottaandPaperclip 3e9e54964c fix(runner): bind ACPX steering to the active provider turn
Preserve flat Runner acknowledgement receipts so native steering emits one
transcript item. Add a pinned Hermes regression across Rust PRP with distinct
durable and provider turn identities and stale-control rejection.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 22:33:24 -05:00
DottaandPaperclip 638296ab2a fix(hermes): retain managed permissions through native restore
Cover native edit approval within the authorized invocation and record bounded,
versioned tool grants per conversation. Recheck policy and cancellation before
each operation. Restore named custom connections through their configured
identity and reject native route fallback. Bind assigned skill contents across
temporary lease replacements while preserving their write protection.

Qualify real native once/deny/session file effects, provider restart, distinct
same-name tool identities, conversation isolation and unanswered permission
cancellation with the credential-free production ACPX fixture. Update the
reviewed platform closure pins and document the remaining release gates.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 21:20:42 -05:00
DottaandPaperclip ef0d855034 test(hermes): enforce native image evidence gate
Require saved image checks to pass after account and billing evidence is collected. Advance the explicit image definition to version 2.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 15:16:21 -05:00
DottaandPaperclip 9478ead982 test(hermes): add native image acceptance journey
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 14:58:47 -05:00
DottaandPaperclip af87a249c1 docs(hermes): record the passing local native Stop proof
Record the committed-source browser cancellation, stale-answer denial, process retirement and reported cost. Keep historical failures and remaining release gates explicit.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 14:13:01 -05:00
DottaandPaperclip 2400ecdb8d test(hermes): require the persisted native cancellation receipt
Bind the expired card to its original unanswered request and require the canonical zero-answer cancellation result. Preserve the failed live measurement and advance qualification definitions.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 14:01:00 -05:00
DottaandPaperclip a14a752ec8 fix(hermes): preserve cancelled usage at the ACPX wire boundary
Bind usage to the admitted native session and turn before cancelled prompt settlement. Check its durable PRP carrier and harden qualification receipt identity and budget capture.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 13:37:58 -05:00
DottaandPaperclip 36125adb4e test(hermes): pin lower qualification campaign budgets
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 12:57:55 -05:00
DottaandPaperclip ec40b1eb32 fix(hermes): retain the owned billing receipt during Stop
Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-08 12:28:13 -05:00