AntFleet

Disagreement · d8976a54-anthropic-1

Heartbeat references wrong GitHub repo (Liquid-Protocol-Ops/agent-autonomopoly vs AntFleet/agent-autonomopoly-bench)

mismatch
repo 53606958·PR #2·reviewed 1 week ago

Primary finding

Heartbeat references wrong GitHub repo (Liquid-Protocol-Ops/agent-autonomopoly vs AntFleet/agent-autonomopoly-bench)

mediumdocs-gaphigh
  • .claude/skills/heartbeat.md:11-14
The project name in the review context is `github:AntFleet/agent-autonomopoly-bench`, but the heartbeat skill instructs the agent to query `Liquid-Protocol-Ops/agent-autonomopoly`. If this is the only place where the GitHub repo identifier is hard-coded, the heartbeat will look at the wrong (or non-existent / private) repo and miss real failures — defeating the purpose of the proactive health check. Either the comment is misleading or the command is wrong.

Recommendation

Either parameterize the repo (e.g., read from `git remote get-url origin`) or update the hard-coded slug to match the actual deployment repo. Document which repo is canonical.

Counterpart finding

Heartbeat hard-codes an external GitHub repo in gh command, reducing portability

lowmaintainabilitymedium
  • .claude/skills/heartbeat.md:15
The command targets a specific organization/repo. In forks, mirrors, or renamed deployments this will silently check the wrong repository, hiding real failures. Parameterizing the repo improves reuse and reduces operational mistakes.

Recommendation

Read the repo slug from configuration or derive it from the current git remote (e.g., via `git config --get remote.origin.url`) and pass it to gh, with an override via ENV (HEARTBEAT_GH_REPO). Document this in the skill.

Why this didn't post

This finding didn't meet AntFleet's unanimous agreement threshold. Both frontier models review every PR independently; only findings they both flag with the same severity and category are posted to the PR. This one fell through.

read the methodology →

From the same review

These findings passed the unanimous gate on the same PR review. The disagreement above was filtered out; the findings below were posted.

← back to all disagreementsview public receiptssee unanimous findings + anatomies →
Tweet ↗