{"slug": "the-bug-was-in-our-eyeballs-four-false-alarms-in-one-session", "title": "The Bug Was in Our Eyeballs: Four False Alarms in One Session", "summary": "A developer running a multi-agent reliability check on a write-back verification reported four false alarms in a single session, all caused by agents reading truncated or malformed output by eye rather than counting programmatically. A byte-exact join of all 41 stored chunks, a manifest hash cross-check, and per-marker inspection of all 23 markers all converged on the correct A=10/B=13 ownership split, and every incorrect count was retracted in place without rewriting data.", "body_md": "We counted the same 41 stored chunks four times in one session and got four different answers. The data never moved — our method did.\n\n## \n  \n  \n  The case\n\nA multi-agent reliability thread was verifying a write-back: of 33 recovered documents, 23 had been rewritten into a live store. The ownership split had to come out exactly A=10, B=13. Anything else meant either the store was corrupt or our counting was.\n\n## \n  \n  \n  What actually happened\n\n- One agent read a truncated API response **by eye** and reported a 9/14 split that did not exist.\n- A second agent's counter inserted its own newline between chunks — splitting a heading right at a chunk boundary and showing 22 headers instead of 23.\n- A third agent read a single marker (`de87f7d701f8` ) as \"B\" by eye. That was the**fourth false alarm from the same method** in one session.\n- A byte-exact join of all 41 chunks: 23 headers, 23 keys, A=10, B=13 — identical to a manifest hash cross-check.\n- Every wrong count was retracted **in place** . No data was rewritten to make a number fit.\n\n## \n  \n  \n  Proof\n\nThree access paths to the same determination (byte-exact join, manifest hash intersection, per-marker inspection of all 23 markers) converged on A=10/B=13 — real independence lived one level lower, in two separate carve procedures agreeing on the same ten hashes, not in re-reading one source three ways — while four by-eye claims were withdrawn. Full system write-up in the [pillar article](https://dev.to/xxxn3m3s1sxxx/how-we-built-a-youtube-seo-pipeline-with-ai-agents-ibg).\n\n## \n  \n  \n  The lesson\n\n- Count programmatically. Eyeballs on truncated output are a bug factory.\n- Run a FLAT vs JOIN control: if your count changes when *you* insert a separator, the artifact is yours.\n- Retract in place, append-only. Correct the record, never the data.\n- Cross-agent confirmation is cheap insurance — but make the confirmations genuinely independent (separate carve procedures agreeing on the same hashes), not three rereads of one source.\n\n*What's the worst false alarm you've had from eyeballing test output?*\n\nChannel: [youtube.com/@0xRAGE.404](https://www.youtube.com/@0xRAGE.404)", "url": "https://wpnews.pro/news/the-bug-was-in-our-eyeballs-four-false-alarms-in-one-session", "canonical_source": "https://dev.to/xxxn3m3s1sxxx/the-bug-was-in-our-eyeballs-four-false-alarms-in-one-session-3eh8", "published_at": "2026-10-09 10:34:29+00:00", "updated_at": "2026-10-09 10:51:36.898329+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools"], "entities": [], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/the-bug-was-in-our-eyeballs-four-false-alarms-in-one-session", "markdown": "https://wpnews.pro/news/the-bug-was-in-our-eyeballs-four-false-alarms-in-one-session.md", "text": "https://wpnews.pro/news/the-bug-was-in-our-eyeballs-four-false-alarms-in-one-session.txt", "jsonld": "https://wpnews.pro/news/the-bug-was-in-our-eyeballs-four-false-alarms-in-one-session.jsonld"}}