mondo report for 2026-05-07

[github-actions]

Report date: 2026-05-07
Lookback window: 2026-04-23 – 2026-05-07 (14 days)

Agent activity in monarch-initiative/mondo is minimal for this period. 1 agent-touched item was detected, with 1 agent summon (a human referencing Claude Code in a comment), and 0 agent-authored PRs or issues. No merged or closed agent work was found.

[github-actions]

Mondo Agent Activity Review — 2026-05-07

What seems to be working

• Human-led use of Claude Code as a tool is happening organically. In issue #10159 (OMIA-Mondo alignment), contributor katiermullen explicitly noted using Claude Code to search for Mondo terms missing gene relations from the OMIA database. This is productive AI-assisted curation, even though no autonomous agent authored any PR.
• The OMIA alignment work in that issue is substantive and ongoing, with detailed process summaries posted over several weeks — suggesting Claude Code is helping with real, complex curation tasks when invoked by a human.

What seems not to be working

• No autonomous agent activity. Zero agent-authored PRs or issues in the 14-day window. No summons led to a follow-up agent action — the single mention in issue #10159 describes past use rather than an active delegation.
• No agent-driven issue triage or synonym/label edits, which are among the most routine, bounded tasks in a disease ontology workflow.

Recurring failure modes or friction

• The pattern visible here — a human using Claude Code interactively but not delegating autonomous follow-through — is the dominant mode. This is valuable but doesn't generate observable agent-authored artifacts.
• With only 1 agent-touched item out of 20 scanned in the window, the repo is not yet configured or habituated to invoke agents for routine maintenance work.

Potential missed opportunities

The dossier surfaces 5 open issues with no agent involvement. Three are plausible, bounded candidates:

1. Issue #10237 — synonym request for MONDO:0032730 (leukodystrophy, hypomyelinating, 18)
   The reporter asks for specific synonyms (degs1-related hypomyelinating leukodystrophy, DEGS1-HLD) and suggests a label swap. This is a well-scoped, single-term synonym addition — exactly the kind of edit that can be expressed as an OWL/OBO patch. A confident recommendation: the ask is concrete, the term ID is given, and there is no ambiguity about what should change.

2. Issue #10236 — label correction for MONDO:0030714 (osteogenesis imperfecta, IIA 22)
   The reporter requests a label update (osteogenesis imperfecta type IIA 22 or similar). The issue body supplies the term ID and a suggested new label. This is a straightforward label/synonym edit, well-bounded for an agent PR.

3. Issue #10159 — OMIA-Mondo alignment
   Already partially assisted by Claude Code interactively. The ongoing work (Steps 2B through remaining phases) involves structured pipeline output files. An agent could help validate or apply the remaining ROBOT template outputs, though this is more complex than the synonym requests above and would need katiermullen to delegate explicitly.

Issue #5726 (VeNom alignment) and issue #10230 (disambiguation review) were surfaced in the dossier but are not strong candidates: the VeNom work is being actively driven by katiermullen with evolving plans, and the disambiguation issue is primarily a human curatorial judgment call.

Summary

Agent presence in this window amounts to a single tool-use mention by a human curator. The infrastructure for autonomous agent contributions is not yet active in this repo, but the human team is comfortable using Claude Code interactively. The synonym and label correction issues (#10237, #10236) represent the lowest-friction starting points for an agent-authored PR.