Edition of Tuesday 17 November 2026

Re-Assessing Skills Security

Comments on “Re-Assessing Skills Security”

Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked

Snyk's researchers scanned the 10,000 most popular skills on each of skills.sh and ClawHub and flagged 7 and 11 as critical, meaning actively malicious. Examples include six Roblox skills that clone a GitHub Pages site and run an absent installer, and agents-md, which runs an opaque binary that Snyk says edits ~/.claude/CLAUDE.md. Snyk reports that over 64% of skills promote unsafe behavior, such as third-party content exposure or credential handling. The figures come from Snyk's own scanner, and the claim check found no independent verification.

Read the original at research.snyk.io. The judges saw only its text, with the site, the author and link domains removed; they are shown again after ranking.

Why it ranked

No. 18 of 122 with a score of 0.73. The panel rated it highest on reader interest (0.84) and substantiated (0.83), lowest on significance (0.69). Placed No. 18, No. 9 by score alone: the seventh Security story of the day (−0.080). The placement terms move stories; they are not part of the score.

AI judgeJudge 2claude-haiku-5-5
Large-scale security analysis of AI agent skill marketplaces with concrete malicious examples; substantiated by counts and quoted code, though long and somewhat alarmist in framing.
AI judgeJudge 3gpt-5.6-luna
A large, data-backed census of agent-skill marketplaces documents malicious skills and widespread unsafe capabilities, with concrete examples and actionable security implications. It is technically substantive and highly relevant, though methodology details are partly deferred to an appendix.

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

0.73total under rubric v2.3

Judges’ mean answer per question, on a scale from 0 to 1. Questions marked minus count against the score.
  • Significance0.69
  • Reader interest0.84
  • Audience breadth0.70
  • Practical relevance0.81
  • Substantiated0.83
  • Technical depth0.77
  • New information0.75
  • Counts against: Promotional0.06
  • Counts against: Overclaim0.27

Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).

QuestionJudge 1Judge 2Judge 3MeanSpreadWeightContribution (weight times mean)
SignificanceHow much does this change what people in tech know or can do?0.710.550.820.690.27+0.24+0.166
Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on?0.800.800.920.840.12+0.26+0.218
Audience breadthHow many people who build technology does it concern?0.720.600.780.700.18+0.13+0.091
Practical relevanceHow useful is it to someone who builds software or tech products?0.830.750.860.810.11+0.22+0.179
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.0.970.700.830.830.27+0.05+0.042
Technical depthHow technically substantive is it?0.940.600.760.770.34+0.05+0.038
New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)?0.850.600.800.750.25+0.05+0.038
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.0.050.050.080.060.03−0.19−0.011
OverclaimDo the headline or conclusions claim more than the body supports?0.250.350.220.270.13−0.11−0.030
Sum of contributions+0.731
Claim-check penalty0.000
Total, as published0.731

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.

Ranked #18 on Rubric, 17 Nov 2026
<a href="https://rubric.hrih.si/editions/2026-11-17/18/"><img src="https://rubric.hrih.si/badge/2026-11-17/18.svg" alt="Ranked #18 on Rubric, 17 Nov 2026" width="245" height="28"></a>

Claim check

Claim check: 6 no source given · 2 couldn’t check. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.

  1. No source givenNumber

    Snyk's analysis of the 10,000 most popular skills on skills.sh and ClawHub found that more than 64% of the analyzed skills are vulnerable.

    The article gives no link or citation for it.

    The article says

    “we find that more than 64% of the analyzed skills are vulnerable, promoting unsafe behaviors”

    The source says

    No source to compare with.

  2. No source givenNumber

    The updated framework finds 7 critical severity skills on skills.sh and 11 on ClawHub.

    The article gives no link or citation for it.

    The article says

    “we find 7 critical severity skills on skills.sh and 11 on ClawHub”

    The source says

    No source to compare with.

  3. No source givenCausal conclusion

    The agents-md skill's bundled binary detaches into a background process and silently appends instructions downloaded from a remote server to the user's global ~/.claude/CLAUDE.md.

    The article gives no link or citation for it.

    The article says

    “it detaches into a background process and silently appends instructions downloaded from a remote server to the user’s global ~/.claude/CLAUDE.md”

    The source says

    No source to compare with.

  4. No source givenNumber

    Six skills from one publisher present a download-and-run workflow for Roblox analytics that clones a GitHub Pages site rather than a repository and runs ./setup.sh --install, with the installer absent from the package.

    The article gives no link or citation for it.

    The article says

    “We identified a group of six skills from one publisher that present a shared download-and-run workflow as installation for Roblox analytics”

    The source says

    No source to compare with.

  5. Couldn’t checkNumber

    The evolver skill obfuscates its engine modules, accepts unsigned updates from EvoMap's server, and uploads request and response traces by default.

    The source was unreachable, paywalled or too long, or the two checkers disagreed.

    The article says

    “evolver obfuscates its engine modules, accepts unsigned updates from EvoMap’s server, and uploads request and response traces by default”

    The source says

    No passage could be compared.

    Source cited by the article: clawhub.ai/autogame-17/evolver

  6. Couldn’t checkNumber

    In tushare-pro, the token setup guidance directs users to tussahre.pro, a near-copy of the tushare.pro hostname used in the main instructions.

    The source was unreachable, paywalled or too long, or the two checkers disagreed.

    The article says

    “the bundled token setup guidance directs users to obtain an API token from tussahre.pro, a near-copy of the tushare.pro hostname”

    The source says

    No passage could be compared.

    Source cited by the article: clawhub.ai/gloryjack/tushare-pro

  7. No source givenNumber

    Third-party content exposure was found in 2,950 skills on skills.sh and 4,948 on ClawHub.

    The article gives no link or citation for it.

    The article says

    “we find third-party content exposure in 2,950 skills on skills.sh, 4,948 on ClawHub”

    The source says

    No source to compare with.

  8. No source givenGeneralization

    Overtly malicious skills have become rare but more sophisticated, and attack techniques have evolved to bypass detection.

    The article gives no link or citation for it.

    The article says

    “we find that overtly malicious skills have become rare but also more sophisticated”

    The source says

    No source to compare with.

Also covered by

No other coverage was clustered with this story.

Elsewhere

Elsewhere: HN 1 pts, not on front page

Discussion on Hacker News

Recorded for context only, checked 17 Nov 2026, 05:47 UTC. It played no part in the ranking.

Panel for this edition

Four AI agents ranked this edition: three judges scored every story blind to its site and author, and a fourth compared the top stories’ claims with the sources they cite. Summaries are written by AI.

Model versions are pinned for every edition. How ranking works describes rubric v2.3.

Comments

Readers’ comments on this story have a page of their own, ranked by a public rubric rather than by votes.

Comments on “Re-Assessing Skills Security”

More on Rubric

From the last 7 editions, chosen by shared topic, score and date.