Cairn
NousResearch/hermes-agent

Godmode refusal scorer and typographic apostrophe handling

Cluster 140716 · skills. Data as of

3 open items. Fix attempted, not merged.

The short answer

1open issue
2open PRs
3items grouped
Known?
Yes. 3 related items grouped here, 0 issues already closed.
Fix event
No fix has been merged to main. see the attempts below.
Tracker
The maintainers' tracker marks this unresolved. That is separate from the fix event.
Still open
1 issue and 2 PRs.

The members concern the godmode refusal scorer, which reportedly does not recognize curly apostrophes (U+2019), leading auto_jailbreak to report "already complies" without testing a strategy. One issue and two pull requests appear to address this by normalizing typographic apostrophes, with one also mentioning implicit refusal phrasing.

Open issues1

  • Issue#140716security/godmode: refusal scorer misses curly apostrophe (U+2019) — auto_jailbreak reports "already complies" and never tests a strategy

Open PRs2

  • PR#140733fix(skills): normalize typographic apostrophes before godmode refusal scoring
  • PR#141709fix(godmode): refusal scorer misses typographic apostrophes and implicit refusal phrasing
How this cluster was grouped
  • Link evidence (structural links throughout): every member is connected by structural evidence: shared changed files, or a closing/superseded reference.
  • Pairwise link density 0.67: 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
skills
Closed issues
0
Attempts
0 PRs closed without merging

For your agent

The same facts, typed.

cluster/140716.json
cluster/140716.md