All workflows

Workflow runs

• Hermes integration: tool results escape non-ASCII text as \uXXXX (~3x tokens for CJK)

  completed successfully:
  Triage labels#1309:Issue#6561openedbyyqm-307
  Nov 14, 202618s
  More actions for workflow run 1309

• Reflect loop turns hard-clamp HINDSIGHT_API_REFLECT_MAX_COMPLETION_TOKENS to 1024, reintroducing #4437 truncation

  completed successfully:
  Triage labels#1308:Issue#6560openedbyERroooooR
  Nov 14, 202617s
  More actions for workflow run 1308

• Recover pending DeepSeek Harness retains after a desktop restart

  completed successfully:
  Triage labels#1307:Issue#6559openedbyijevin
  Nov 14, 202623s
  More actions for workflow run 1307

• fix(coding-agents, reflect): agent memory keeps the lessons a session learned

  failed:
  CI#10801:Pull request#6558openedbynicoloboschi
  exp/reflect-clbench

  Nov 14, 202613m 26s
  More actions for workflow run 10801

• feat(knowledge-banks): search baseline for benchmarks — keywords in every language, capped results, search trace, reflect on the real search

  completed successfully:
  CI#10800:Pull request#6542synchronizebynicoloboschi
  knowledge-reflect-search

  Nov 14, 202626m 4s
  More actions for workflow run 10800

• coding-agents: background jobs spawn bare node, fail with ENOENT when node is not on the hook's PATH

  completed successfully:
  Triage labels#1306:Issue#6557openedbykfiz
  Nov 14, 202616s
  More actions for workflow run 1306

• Windows Smoke Test

  completed successfully:
  Windows Smoke Test#169:Scheduled
  main

  Nov 14, 202610m 0s
  More actions for workflow run 169

• Performance Tests

  failed:
  Performance Tests#229:Scheduled
  main

  Nov 14, 20261h 0m 20s
  More actions for workflow run 229

• Security Scan

  failed:
  Security Scan#75:Scheduled
  main

  Nov 14, 20266m 58s
  More actions for workflow run 75

• coding-agents: Codex reader buffers discarded tool results; large-session Stop hook aborts with SIGABRT

  completed successfully:
  Triage labels#1305:Issue#6556openedbyjohnnynalley
  Nov 14, 202616s
  More actions for workflow run 1305

• Decision-orchestrated reflect loops to max_iterations: sufficiency check reads only the head of the evidence, follow-up rounds re-fetch the same pages

  completed successfully:
  Triage labels#1304:Issue#6555openedbyjervaise
  Nov 14, 202617s
  More actions for workflow run 1304

• Decision-model calls skip per-bank attribution (no user field, no X-Hindsight-Bank-Id header)

  completed successfully:
  Triage labels#1303:Issue#6554openedbyFedotCompot
  Nov 14, 202615s
  More actions for workflow run 1303

• coding-agents: with the server unreachable, SessionStart takes 21 s and every UserPromptSubmit 16 s, even with reflectTimeoutMs and injectTimeoutMs at 3 s

  completed successfully:
  Triage labels#1302:Issue#6553openedbyichoosetoaccept
  Nov 13, 202616s
  More actions for workflow run 1302

• reflect fails with HTTP 400 on DeepSeek when configured as LLM_PROVIDER=openai with the DeepSeek base URL

  completed successfully:
  Triage labels#1301:Issue#6552openedbyvernmic
  Nov 13, 202617s
  More actions for workflow run 1301

• update_mode="append" without a tags field clears the document's tags and every memory's tags

  completed successfully:
  Triage labels#1300:Issue#6551openedbyvernmic
  Nov 13, 202618s
  More actions for workflow run 1300

• OpenAPI: GET and PATCH /memories/{memory_id} declare an empty 200 schema

  completed successfully:
  Triage labels#1299:Issue#6550openedbyetzelm
  Nov 13, 202617s
  More actions for workflow run 1299

• Mental models: preview a changed source_query with dry-run-refresh before saving it

  completed successfully:
  Triage labels#1298:Issue#6549openedbyetzelm
  Nov 13, 202615s
  More actions for workflow run 1298

• [hermes integration] _pending_retain_ops grows for the whole session when prefetch_waits_for_retain is false

  completed successfully:
  Triage labels#1297:Issue#6548openedbyjransom87
  Nov 13, 202617s
  More actions for workflow run 1297

• Link background operations and LLM requests in both directions

  completed successfully:
  Triage labels#1296:Issue#6547openedbyFedotCompot
  Nov 13, 202617s
  More actions for workflow run 1296

• LLM Requests chart: plot time spent in calls and failed calls, overall, per model and per operation

  completed successfully:
  Triage labels#1295:Issue#6546openedbyFedotCompot
  Nov 13, 202618s
  More actions for workflow run 1295

• Background operations: show when an operation started, when it ended, and how long it ran

  completed successfully:
  Triage labels#1294:Issue#6545openedbyFedotCompot
  Nov 13, 202616s
  More actions for workflow run 1294

• Control plane: running total switch on the "Memories by ingested time" chart

  completed successfully:
  Triage labels#1293:Issue#6544openedbyFedotCompot
  Nov 13, 202617s
  More actions for workflow run 1293

• Correlate model-call history with the active distributed trace

  completed successfully:
  Triage labels#1292:Issue#6543openedbymz2
  Nov 13, 202620s
  More actions for workflow run 1292

• feat(knowledge-banks): search baseline for benchmarks — keywords in every language, capped results, search trace, reflect on the real search

  cancelled:
  CI#10799:Pull request#6542openedbynicoloboschi
  knowledge-reflect-search

  Nov 13, 20265m 45s
  More actions for workflow run 10799

• Link consolidation traces to the requests that supplied their memories

  completed successfully:
  Triage labels#1291:Issue#6541openedbymz2
  Nov 13, 202625s
  More actions for workflow run 1291