Cairn

Kanban protocol-violation on clean worker exit handling

NousResearch/hermes-agent · cluster 27178 · cron

Data as of

3 open items. Fix merged in #112618. Check it covers your case.

The short answer

3open issues
0open PRs
13items grouped
Known?
Yes. 13 related items grouped here, 5 issues already closed.
Fix event
A fix was merged to main in PR #112618. Check it covers your case. No release data is tracked.
Tracker
The maintainers' tracker marks this unresolved. That is separate from the fix event.
Still open
3 issues and 0 PRs.

Multiple bugs, fixes, and features addressing how the Kanban dispatcher handles workers that exit with rc=0 without calling kanban_complete or kanban_block, which incorrectly triggers protocol_violation errors, blocks tasks, and buries real errors. Members cover detection, retry logic, error surfacing, and downstream pipeline impact.

Open issues3

  • Issue#27178Kanban worker reports 'protocol_violation' when agent ends turn with text response instead of calling kanban_complete/kanban_block
  • Issue#27924[Bug]: Clean-exit protocol violation blocks downstream pipeline tasks when workers complete silently
  • Issue#63504Kanban dan-pr worker crashes/protocol violation block JTAPI show-tech fix

Open PRs0

No open pull requests.

Fixes on record

Merged to main. A stamp counts the issues a fix is linked to.

  • #112618 Kanban board shows the worker's own last output on protocol violations and crashes (#88603, #46593, salvage #88815)

    merged to main

    files: hermes_cli/kanban_db_dispatch.py, tests/hermes_cli/test_kanban_core_functionality.py, website/docs/user-guide/features/kanban.md

4 PRs closed without merging

From the close comments: done elsewhere 3 · other 1

  • #34193 fix(kanban): checkpoint reminder before worker exits with text response (closes #27178) · other → #34217, #1
  • #34324 fix(cron): make protocol_violation blocks sticky to break respawn loop · done elsewhere
  • #46985 fix(kanban): surface real worker error instead of bare "protocol violation" · done elsewhere → 2cfb655, #112618, #88815, #88603
  • #59411 [codex] block unclosed Kanban workers and preserve gave_up · done elsewhere → 452861f
Closed issues · 5
  • Closed#32747Protocol violations (rc=0 without complete/block) don't increment failure_limit counter
  • Closed#44812Kanban dispatcher misclassifies rc=0 worker exits as protocol_violation (new in-the-wild data, 270 events / 160 blocked tasks)
  • Closed#46593kanban: worker exits rc=0 without calling `kanban_complete` shows unhelpful 'protocol violation' — real error buried in logs
  • Closed#46889kanban: worker exits rc=0 without calling `kanban_complete` shows unhelpful 'protocol violation' — real error buried in logs
  • Closed#47893dispatcher treats clean worker exit (rc=0) without kanban_complete/kanban_block as protocol_violation and reverts done → blocked
How this cluster was grouped
  • Link evidence (partly structural links): some members are connected by structural evidence (shared changed files, or a closing/superseded reference); the rest are joined by an automated relevance judgement over their text.
  • Members outside that structural core may belong to a different underlying problem.
  • Pairwise link density 0.11: the share of item pairs that carry a link, counting a structural link in full and a relevance-only link at 0.36. 1.00 means every pair is linked.

Pick up a Lab task

Check whether a fix covers the open issues here, with your own agent. Bring back evidence for maintainers.

Open the Lab

At a glance

Subsystem
cron
Merged PRs
#112618
Closed issues
5
Attempts
4 PRs closed without merging
Fix link checked
16 Sep 2026

For your agent

The same facts, typed.

cluster/27178.json
cluster/27178.md