RepoMedic skill flagged as suspicious despite benign scans (requesting review + guidance) #125

Closed
opened 2026-02-15 17:15:49 -05:00 by yindo · 1 comment
Owner

Originally created by @mrummler17 on GitHub (Feb 11, 2026).

Hi team,

Our ClawHub skill repomedic is currently flagged with “suspicious patterns detected”, but both scanner outputs show benign results:

OpenClaw scan: Benign (high confidence)
VirusTotal: Benign

This skill is a repo-maintenance tool (Dependabot triage, lockfile repair, scoped dependency PRs). It requires GitHub + filesystem access because those are necessary for the intended functionality.

We believe this may be a heuristic false-positive caused by privileged-but-expected repo automation patterns.

To improve clarity on our side, we are publishing an update that adds:

explicit listing summary
required binaries metadata (git, gh, pnpm)
explicit guardrails (no credential exfiltration, no unrelated service access, no destructive actions without explicit user direction)

Could you please review the flag status and advise what additional metadata/signals would help avoid this false-positive classification for legitimate maintenance skills?

Thanks Marcus

Originally created by @mrummler17 on GitHub (Feb 11, 2026). Hi team, Our ClawHub skill repomedic is currently flagged with “suspicious patterns detected”, but both scanner outputs show benign results: OpenClaw scan: Benign (high confidence) VirusTotal: Benign This skill is a repo-maintenance tool (Dependabot triage, lockfile repair, scoped dependency PRs). It requires GitHub + filesystem access because those are necessary for the intended functionality. We believe this may be a heuristic false-positive caused by privileged-but-expected repo automation patterns. To improve clarity on our side, we are publishing an update that adds: explicit listing summary required binaries metadata (git, gh, pnpm) explicit guardrails (no credential exfiltration, no unrelated service access, no destructive actions without explicit user direction) Could you please review the flag status and advise what additional metadata/signals would help avoid this false-positive classification for legitimate maintenance skills? Thanks Marcus
yindo closed this issue 2026-02-15 17:15:49 -05:00
Author
Owner

@mrummler17 commented on GitHub (Feb 13, 2026):

I've updated RepoMedic and this appears to have resolved the issues flagged.

@mrummler17 commented on GitHub (Feb 13, 2026): I've updated RepoMedic and this appears to have resolved the issues flagged.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: openclaw/clawhub#125