NousResearch/hermes-agent

Memory replace ambiguity causing duplicate entries on fallback add

Cluster 60089 · agent. Data as of

2 open items. Fix attempted, not merged.

The short answer

1open issue
1open PR
10items grouped
Known?
Yes. 10 related items grouped here, 0 issues already closed.
Fix event
No fix has been merged to main. 8 PRs closed without merging; see the attempts below.
Tracker
The maintainers' tracker marks this unresolved. That is separate from the fix event.
Still open
1 issue and 1 PR.

These members concern the memory replace operation: when a replace matches multiple entries, a fallback add can create duplicate entries without any signal. One issue reports the problem, and the pull requests propose detecting existing or duplicate new content during ambiguous replace, or improving the resulting error.

Open issues1

  • Issue#60089[Bug]: memory replace → "Multiple entries matched" → fallback add creates duplicate entries with no signal

Open PRs1

  • PR#71819fix(memory): disambiguate replace when new_content already exists

8 PRs closed without merging

From the close comments: stale or abandoned 4 · duplicate 3 · done elsewhere 1

  • #60090 fix(memory): guard against duplicate-entry trap on ambiguous replace · done elsewhere → #60089
  • #60093 fix(memory): detect duplicate new_content in ambiguous replace (#60089) · stale or abandoned
  • #60147 fix(memory): detect duplicate new_content in ambiguous replace (#60089) · stale or abandoned
  • #60208 fix(memory): check for existing content on ambiguous replace to prevent duplicate add (#60089) · duplicate → #60090, #60093, #60147, #60089
  • #60216 fix(memory): detect duplicate content in ambiguous replace errors (#60089) · stale or abandoned
Show 3 more
  • #60321 fix(memory): guard against duplicate-entry trap on ambiguous replace · duplicate → #60090, #60089, #60093, #60147, #60208, #60216
  • #60511 fix(memory): better error when replace() content already exists verbatim (#60089) · stale or abandoned
  • #66916 fix(memory): detect existing-duplicate in ambiguous replace to break silent duplicate loop (#60089) · duplicate → #60208
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(structural links throughout): every member is connected by structural evidence: shared changed files, or a closing/superseded reference.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 0.20: 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
Closed issues
0
Attempts
8 PRs closed without merging

For your agent

The same facts, typed.