The tokens in your agent harness
Comments on “The tokens in your agent harness”
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
An author who read the open-source Codex harness in Rust catalogs the hidden LLM calls it makes beyond the main task: automatic thread naming, reasoning-based activity status (no separate request), auto recaps, memory extraction and consolidation, and a background Guardian security reviewer.
Why it ranked
No. 25 of 43 with a score of 0.60. The panel rated it highest on new information (0.83) and technical depth (0.75), lowest on significance (0.47). Placed No. 25, No. 21 by score alone: the twelfth AI & machine learning story of the day (−0.080). The placement terms move stories; they are not part of the score.
- AI judgeJudge 2claude-haiku-5-5
- Source-level walkthrough of where Codex spends LLM tokens beyond user work; useful to agent builders and well-detailed, though based on code reading rather than measurements and of moderate significance.
- AI judgeJudge 3gpt-5.6-luna
- A technically specific investigation into hidden LLM calls, token usage, recaps, memories, and safety review in an agent harness. It is actionable and novel, though evidence is described through links and some model details may be unstable.
Each judge wrote its line after answering the rubric, with the site and the author still hidden. Every answer behind this is in the table under Score.
Score
0.60total under rubric v2.3
- Significance0.47
- Reader interest0.73
- Audience breadth0.54
- Practical relevance0.69
- Substantiated0.66
- Technical depth0.75
- New information0.83
- Counts against: Promotional0.05
- Counts against: Overclaim0.22
Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).
| Question | Judge 1 | Judge 2 | Judge 3 | Mean | Spread | Weight | Contribution (weight times mean) |
|---|---|---|---|---|---|---|---|
| SignificanceHow much does this change what people in tech know or can do? | 0.50 | 0.35 | 0.55 | 0.47 | 0.20 | +0.24 | +0.112 |
| Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on? | 0.65 | 0.72 | 0.82 | 0.73 | 0.17 | +0.26 | +0.190 |
| Audience breadthHow many people who build technology does it concern? | 0.54 | 0.60 | 0.48 | 0.54 | 0.12 | +0.13 | +0.070 |
| Practical relevanceHow useful is it to someone who builds software or tech products? | 0.75 | 0.65 | 0.68 | 0.69 | 0.10 | +0.22 | +0.153 |
| SubstantiatedHow well is it backed? Consider whether it is the primary source, and whether its main claims come with data, code, a demo, a reproducible method or citations. | 0.65 | 0.60 | 0.72 | 0.66 | 0.12 | +0.05 | +0.033 |
| Technical depthHow technically substantive is it? | 0.85 | 0.70 | 0.70 | 0.75 | 0.15 | +0.05 | +0.038 |
| New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)? | 0.86 | 0.80 | 0.84 | 0.83 | 0.06 | +0.05 | +0.042 |
| PromotionalIs it mainly marketing with little substance (a press release, ad or sponsored post with no real information)? A launch post that carries real news is not promotional. | 0.05 | 0.05 | 0.04 | 0.05 | 0.01 | −0.19 | −0.009 |
| OverclaimDo the headline or conclusions claim more than the body supports? | 0.18 | 0.20 | 0.28 | 0.22 | 0.10 | −0.11 | −0.024 |
| Sum of contributions | +0.603 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.603 | ||||||
Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.
The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.
<a href="https://rubric.hrih.si/editions/2026-11-02/25/"><img src="https://rubric.hrih.si/badge/2026-11-02/25.svg" alt="Ranked #25 on Rubric, 2 Nov 2026" width="238" height="28"></a>Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN 1 pts, not on front page
Recorded for context only, checked 2 Nov 2026, 05:46 UTC. It played no part in the ranking.