31 Commits

Author SHA1 Message Date
amos
913d3c367f Add the usable task catalog and admit batch 6 tasks
All 161 task packages, including the 21 batch 6 C++17 tasks that passed
gates 1/2, are now in Git. rollout.catalog builds docs/task-catalog.{json,md}
from the task packages, a fresh trace5 calibration (128/128 PASS), every
verified controller receipt on the current task digest, documented holds
and digest-bound difficulty reviews. 123 tasks are usable (6 HARD),
5 are on hold and 33 still need a subprocess-only grader.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-18 17:43:15 +08:00
Codex
75408399b6 Add four C++ reliability tasks with verified Qwen screening 2026-10-18 00:26:04 +08:00
amos
b71d605d81 Lower the shared-disk floor to 30 GiB at the user's request
The other users of /home left 57 GiB free; the user chose to keep running
and stop only at 30 GiB. Batch start and runtime disk_low now share that floor.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-17 13:11:21 +08:00
amos
a362d21163 Record batch 6 disk preflight block
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-17 13:00:12 +08:00
Codex
c0a29c489a Record batch5 Qwen evidence and verified cold migration 2026-10-17 12:53:02 +08:00
Codex
d10f062bab Add embedded reliability batch 5 group 4 (five calibrated tasks) 2026-10-17 10:25:22 +08:00
Codex
7559f1a785 Add embedded reliability batch 5 group 3 (five calibrated tasks) 2026-10-17 10:20:18 +08:00
Codex
9e8091320a Add embedded reliability batch 5 group 2 (five calibrated tasks) 2026-10-17 10:18:48 +08:00
Codex
205118d8c4 Add embedded reliability batch 5 group 1 (five calibrated tasks) 2026-10-17 10:17:46 +08:00
Codex
1f58ebdb6b Record Qwen refill results and verified cold migration 2026-10-17 10:04:29 +08:00
Codex
468dbe68a2 Add five symptom-based embedded reliability tasks and Qwen pilots 2026-10-17 03:27:57 +08:00
Codex
30762e4d29 docs: allow evidence-based symptom task prompts 2026-10-17 01:39:03 +08:00
amos
05bd1e10c3 Re-author three blocked Bos graders so candidates only run in a subprocess
S500_n3_133 v3, S500_n3_102 v4 and S500_a2_426889b9a968 v3 move every
expected value and assertion into check.py; the candidate process only
receives inputs and prints raw observations. isolation-blocked.txt and the
in-process tests.py are removed and each task gains cheat_read_grader.
Formal trace5 calibration on 36 passed with these exact task digests.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-17 01:32:04 +08:00
Codex
fe65c3bf99 Add 20 HWE repair tasks and record blocked GLM pilots 2026-10-17 01:29:59 +08:00
Codex
69521e54e0 Add 15 embedded repair tasks and correct grader-read controls
Complete batch 20261016-C-embedded-2: 45 verified GLM/Pi samples, independent regrades and archives; preserve difficulty misses and Salty blockage.
2026-10-16 14:23:48 +08:00
amos
6d6ae4f257 Add five C11 serial protocol repair pilot tasks 2026-10-15 22:13:48 +08:00
amos
705ec37c20 docs: record the Bos isolation fix and its recalibration
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-15 12:49:22 +08:00
amos
f1acc8c938 Import Bos direct batch 20261014-A-bos-direct 2026-10-15 00:37:44 +08:00
amos
3525e76b9a docs: record the 36 working-copy migration
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-14 17:24:54 +08:00
amos
293c2c7a8e docs: task-making spec, Codex factory brief and storage layout
Rewrite task-authoring.md as one spec (four source routes, exclusions,
prompt/grader/fixture rules, four gates, difficulty pools), add the
brief that gives Codex the machines, storage layout, data map, commands
and red lines, and record the Bos-qiao raw-data audit.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-14 15:22:36 +08:00
amos
c6435255f9 docs: record the user-specified Git remote
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 20:49:41 +08:00
amos
94065e75e0 docs: record the architecture fixes, trace5 and the evdev correction
Also keeps the parallel session's note on locally expanded trace samples.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 20:41:18 +08:00
amos
71b9cc4730 docs: record the Qwen 20-task baseline run
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 20:04:09 +08:00
amos
e3002bd5b2 docs: record the M1 twenty-task pilot and next steps
40 complete trajectories over 20 Bos-qiao tasks in three families (33
PASS, 7 FAIL), 40/40 consistent regrades, five channel/I/O exclusions
re-attempted after the proxy retry, and the sealed batch verified on
Salty.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 03:30:48 +08:00
amos
7071ecfa7f fix: raise the native trace limit to 64 MiB
Bos tasks with 131072-token reasoning produce ~110 bytes per streamed
token plus a full context per request; the trace3 batch hit 16 MiB
before finishing. Adapter and host share the new limit, and the size
tests now derive their data from MAX_TRACE_BYTES. Runtime trace4 carries
this adapter; all 21 tasks recalibrated.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 23:05:39 +08:00
amos
8b542bd3ed docs: point host 36 runs at the local-disk copy
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 22:39:30 +08:00
amos
1c7e13d6ae docs: record Bos import, first batch failure and trace3 fix
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 22:38:39 +08:00
amos
acfa601a56 feat: multi-file task contract, model table and shared I/O helpers
- Tasks may declare up to 16 candidate files; the whole template becomes
  the Agent workspace while only candidate files are frozen and graded.
- Optional fixtures add more cheats or legal alternatives (five trials
  each); optional grader_files ship private tests next to verify.py.
- Tasks without an argv oracle export call_type=verifier ground truth.
- Upstream model routes live in one MODELS table.
- One bounded no-follow reader, one exclusive writer and one SIGTERM
  helper replace per-module copies; drop unused validate_task.

frame-decoder keeps its task digest, calibration receipts and exported
query. Verified on host 36: internal network probe, 20-trial calibration
and a real GLM rollout (PASS, evidence VERIFIED, cleanup confirmed).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 21:25:09 +08:00
amos
e524d9eda5 docs: deployment map, task authoring standard and Qwen trial record
Add the current and target deployment diagrams, the task author standard
and copyable spec template, the five-field data contract, and record the
Qwen Responses trial and pending-create cleanup fix in progress and
acceptance.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 21:06:06 +08:00
amos
d205c44249 fix: cut agent internet egress and loosen grader timeouts
The dedicated Pi network was a plain bridge, so the Agent could reach the
internet and fetch upstream fixes. Create it with --internal, make run
accept only internal networks, and default to --network none.

Verifier limits of 1s per case and 15s per grade turned host load into
false FAIL verdicts. Use 5s per case, 30s compile, 10s isolation probe and
timeout_seconds=60; frame-decoder task version becomes 2.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 20:30:00 +08:00
amos
a0466b6129 docs: record verified rollout and infrastructure handoff 2026-10-12 11:15:45 +08:00