AntFleet

Disagreement · a255f8b5-anthropic-2

Status page intro contradicts heartbeat's documented behavior on forks

mismatch
repo 6f7fc663·PR #6·reviewed 1 week ago

Primary finding

Status page intro contradicts heartbeat's documented behavior on forks

mediumdocs-gaphigh
  • docs/status.md:12-14
  • docs/_config.yml:5
The status page claims a fork gets a working /status/ with no extra config, but the Jekyll config hard-codes the upstream owner's URL (aaronjmars/aeon). A fork would need to edit _config.yml (url, baseurl, author.url, social_links) to get correct absolute links. The user-facing claim is therefore misleading — it is exactly the kind of 'comment that lies about behavior' the review brief calls out.

Recommendation

Either make _config.yml fork-agnostic (omit url/baseurl so GitHub Pages computes them; use site.github.* metadata) or soften the docs claim to require a one-line edit.

Counterpart finding

Hardcoded schedule string in status template is brittle; can become stale if heartbeat schedule changes

lowdocs-gaphigh
  • docs/status.md:13
The template asserts the heartbeat runs 3× daily at fixed times, but the SKILL instructions generate the page for whatever schedule is configured in aeon.yml. If schedules differ (or change later), the page will mislead readers. This is a documentation fragility that will surface as incorrect operational info.

Recommendation

Remove the hardcoded schedule from the static template and instead write a generic line (e.g., “Auto-generated by the heartbeat skill on its schedule”). The generated page should include a “Next scheduled run:” line computed from aeon.yml per the SKILL rules, ensuring the visible schedule always matches configuration.

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 ↗