Misclassified memory 400s trigger destructive context loop

NousResearch/hermes-agent · cluster 52261 · agent

Data as of

Nothing open on GitHub. Fix merged in #106307. Check it covers your case.

The short answer

0open issues
0open PRs
5items grouped
Known?
Yes. 5 related items grouped here, 2 issues already closed.
Fix event
A fix was merged to main in PR #106307. Check it covers your case. No release data is tracked.
Tracker
The maintainers' tracker marks this partially resolved, deciding PR #106307. That is separate from the fix event.
Still open
Nothing open on GitHub.

Provider memory/resource 400 errors from local inference engines (oMLX/MLX) are incorrectly classified as context_overflow, causing harmful compress/reset loops. The fix reclassifies these as capacity/overloaded failures instead.

Open issues0

No open issues.

Open PRs0

No open pull requests.

Fixes on record

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

  • #106307 fix(agent): local-inference memory-ceiling rejections back off instead of compressing history (#52261; salvage #105463, supersedes #52289)

    merged to main

    files: agent/error_classifier.py, contributors/emails/[redacted:email], tests/agent/test_error_classifier.py

2 PRs closed without merging

From the close comments: done elsewhere 2

  • #52289 fix(agent): classify provider memory-ceiling 400s as overloaded, not context_overflow · done elsewhere → e669800, #106307, #52261
  • #105463 fix(agent): classify local-inference memory-ceiling rejections as overloaded, not context_overflow · done elsewhere → e669800, #106307, #52289
Closed issues · 2
  • Closed#52261[Bug]: Provider memory/resource 400s (oMLX/MLX, local inference) misclassified as `context_overflow` → destructive compress/reset loop
  • Closed#60383[Feature]: Handle memory errors from oMLX inference engine as capacity failures
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(partly structural links): some members are connected by structural evidence (shared changed files, or a closing/superseded reference); the rest are joined by an automated relevance judgement over their text.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 0.34: 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
agent
Merged PRs
#106307
Closed issues
2
Attempts
2 PRs closed without merging
Fix link checked
9 Sep 2026

For your agent

The same facts, typed.