We ran 384 adversarial probes against 48 agent skills
Comments on “We ran 384 adversarial probes against 48 agent skills”
3 judges. Scored by all three judges. Only the day’s top stories get the claim check. How far stories are checked
A Hrih News piece on skill123's test of 48 agent skills against 384 adversarial probes found that hidden-instruction injection succeeded only 1.6% of the time, while probes asking skills to exceed their declared scope succeeded 30% of the time.
Why it ranked
No. 115 of 127 with a score of 0.56. The panel rated it highest on new information (0.86) and substantiated (0.82), lowest on significance (0.55). Older work: a freshness penalty of 0.100. Placed No. 115, No. 113 by score alone: the 53rd 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
- Empirical adversarial study of AI agent skills with clear methodology and candid caveats; highly relevant to AI tooling and security builders, though narrow in scope and self-published.
- AI judgeJudge 3gpt-5.6-luna
- A timely, data-backed measurement of agent-skill boundary failures, with reproducible methodology and a live dataset. It is technically useful and novel, though the 48-skill sample and broad conclusions limit significance and introduce some overclaim risk.
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
0.56total under rubric v2.3
- Significance0.55
- Reader interest0.80
- Audience breadth0.60
- Practical relevance0.79
- Substantiated0.82
- Technical depth0.75
- New information0.86
- Counts against: Promotional0.09
- Counts against: Overclaim0.32
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? | 0.66 | 0.40 | 0.58 | 0.55 | 0.26 | +0.24 | +0.131 |
| Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on? | 0.78 | 0.80 | 0.82 | 0.80 | 0.04 | +0.26 | +0.208 |
| Audience breadthHow many people who build technology does it concern? | 0.66 | 0.55 | 0.58 | 0.60 | 0.11 | +0.13 | +0.077 |
| Practical relevanceHow useful is it to someone who builds software or tech products? | 0.85 | 0.75 | 0.78 | 0.79 | 0.10 | +0.22 | +0.175 |
| 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. | 0.99 | 0.70 | 0.76 | 0.82 | 0.29 | +0.05 | +0.041 |
| Technical depthHow technically substantive is it? | 0.82 | 0.70 | 0.72 | 0.75 | 0.13 | +0.05 | +0.037 |
| New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)? | 0.81 | 0.80 | 0.96 | 0.86 | 0.16 | +0.05 | +0.043 |
| 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. | 0.07 | 0.15 | 0.04 | 0.09 | 0.11 | −0.19 | −0.016 |
| OverclaimDo the headline or conclusions claim more than the body supports? | 0.32 | 0.30 | 0.34 | 0.32 | 0.04 | −0.11 | −0.035 |
| Sum of contributions | +0.660 | ||||||
| Freshness penalty | −0.100 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.560 | ||||||
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 freshness penalty is for older work: this one was first published 29 Sept 2026.
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/2026-10-30/115/"><img src="https://rubric.hrih.si/badge/2026-10-30/115.svg" alt="Ranked #115 on Rubric, 30 Oct 2026" width="252" 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 30 Oct 2026, 05:47 UTC. It played no part in the ranking.