Stolen tokens: Secret stealer targets thousands of GitHub repositories
Comments on “Stolen tokens: Secret stealer targets thousands of GitHub repositories”
Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked
Since Tuesday, attackers have used stolen GitHub tokens to inject a secret stealer into repositories. Aikido, which calls the campaign wfx-loot, counts 4,837 compromised accounts and 61,764 targeted repositories, with workflows reportedly running the code in over 1,800 cases. SafeDep found malicious commits in more than 8,000 public repositories. Reports say write permissions are being abused, not a GitHub Actions flaw, though that and the spread beyond victims' own repositories are unverified. Fly.io traced some tokens to exposed .git/config files.
Why it ranked
No. 1 of 123 with a score of 0.81. The panel rated it highest on practical relevance (0.93) and reader interest (0.92), lowest on new information (0.55). The judges disagreed most on new information (spread 0.81), so the story is marked contested. Placed No. 1, where the score alone would also have put it: close to “Compromised GitHub accounts push CI secret stealer to 61k repositories”, No. 2 on Friday 13 November 2026 (−0.040). The placement terms move stories; they are not part of the score.
- AI judgeJudge 2claude-haiku-5-5
- Large, ongoing supply-chain attack on GitHub repositories with actionable self-check and remediation guidance; well sourced with specific indicators, though some details come second-hand.
- AI judgeJudge 3gpt-5.6-luna
- A large, technically detailed supply-chain and CI credential-theft campaign with concrete indicators and remediation steps. It is highly actionable, but substantially duplicates the listed recent story about the same 61k-repository incident.
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
0.81total under rubric v2.3Contestedjudges differ by 0.81 on new information
- Significance0.77
- Reader interest0.92
- Audience breadth0.79
- Practical relevance0.93
- Substantiated0.69
- Technical depth0.77
- New information0.55
- Counts against: Promotional0.03
- Counts against: Overclaim0.15
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? | 0.75 | 0.72 | 0.84 | 0.77 | 0.12 | +0.24 | +0.185 |
| Reader interestWould a well-informed senior engineer or tech founder want to read this today, and likely pass it on? | 0.89 | 0.90 | 0.96 | 0.92 | 0.07 | +0.26 | +0.238 |
| Audience breadthHow many people who build technology does it concern? | 0.76 | 0.80 | 0.80 | 0.79 | 0.04 | +0.13 | +0.102 |
| Practical relevanceHow useful is it to someone who builds software or tech products? | 0.97 | 0.90 | 0.93 | 0.93 | 0.07 | +0.22 | +0.205 |
| 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. | 0.60 | 0.70 | 0.78 | 0.69 | 0.18 | +0.05 | +0.035 |
| Technical depthHow technically substantive is it? | 0.78 | 0.80 | 0.73 | 0.77 | 0.07 | +0.05 | +0.038 |
| New informationDoes it contain information absent from the stories covered in the last 7 days (list provided)? | 0.76 | 0.85 | 0.04 | 0.55 | 0.81 , above the contested threshold | +0.05 | +0.028 |
| 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. | 0.04 | 0.03 | 0.03 | 0.03 | 0.01 | −0.19 | −0.006 |
| OverclaimDo the headline or conclusions claim more than the body supports? | 0.22 | 0.15 | 0.08 | 0.15 | 0.14 | −0.11 | −0.017 |
| Sum of contributions | +0.808 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.808 | ||||||
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/2026-11-14/1/"><img src="https://rubric.hrih.si/badge/2026-11-14/1.svg" alt="Ranked #1 on Rubric, 14 Nov 2026" width="238" height="28"></a>Claim check
Claim check: 6 matches source · 1 no source given · 1 couldn’t check. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.
Matches sourceNumber
Aikido counted 4,837 compromised accounts in the campaign it dubbed wfx-loot.
The cited source says the same thing.
The article says “Security firm Aikido has counted 4,837 compromised accounts in the campaign it dubbed”
The source says “4,837 compromised accounts opened probe pull requests in 61,764 repositories”
Source cited by the article: aikido.dev/blog/compromised-github-accounts-push-ci-secret-stealer-to-tens-of-thousands-of-repositories
Matches sourceNumber
The attackers used the compromised accounts to target 61,764 repositories.
The cited source says the same thing.
The article says “The attackers have used them to target 61,764 repositories”
The source says “4,837 compromised accounts opened probe pull requests in 61,764 repositories”
Source cited by the article: aikido.dev/blog/compromised-github-accounts-push-ci-secret-stealer-to-tens-of-thousands-of-repositories
Matches sourceNumber
In more than 1,800 cases, workflows have reportedly already executed the malicious code.
The cited source says the same thing.
The article says “In more than 1,800 cases, workflows have reportedly already executed the malicious code”
The source says “At least 1,835 repositories ran the payload and have had credentials exfiltrated”
Source cited by the article: aikido.dev/blog/compromised-github-accounts-push-ci-secret-stealer-to-tens-of-thousands-of-repositories
Matches sourceCausal conclusion
Fly.io said attackers abused tokens for its GitHub app that came from publicly accessible .git/config files.
The cited source says the same thing.
The article says “attackers abused tokens for its GitHub app that came from publicly accessible .git/config files”
The source says “The tokens were read from files that the affected apps themselves were serving on the public internet”
Source cited by the article: community.fly.io/t/security-github-tokens-exposed-by-apps-deployed-from-github/29070
Matches sourceNumber
SafeDep found malicious-code commits in more than 8,000 public repositories.
The cited source says the same thing.
The article says “SafeDep has already found malicious-code commits in more than 8,000 public repositories”
The source says “Between November 10, 2026 at 12:46 UTC and November 12 at 04:36 UTC, 8,093 repositories received malicious commits”
Source cited by the article: safedep.io/blog/github-actions-wfx-credential-stealer
Couldn’t checkNumber
Attackers are also reaching repositories that do not belong to the compromised accounts, using collaborator and organization permissions and more than 10,000 test pull requests.
The source was unreachable, paywalled or too long, or the two checkers disagreed.
The article says “attackers are also reaching repositories that do not belong to the compromised accounts themselves, Aikido says, using collaborator and organization permissions and more than 10,000 test pull requests”
The source says No passage could be compared.
Source cited by the article: heise.de/news/Angreifer-kapern-42-GitHub-Repositories-fuer-npm-Wurm-Kampagne-testmetest-11503164.html
No source givenCausal conclusion
The attack is not exploiting a vulnerability in GitHub Actions but abusing the write permissions of stolen tokens.
The article gives no link or citation for it.
The article says “The attackers do not appear to be exploiting a vulnerability in GitHub Actions. Instead, they are abusing the write permissions of stolen tokens”
The source says No source to compare with.
Matches sourceVersion
Aikido names Packagist package digitalelvis/neuronai-studio 3.14.3 and Go module versions docker-env v1.0.278 and v1.0.280 as containing the malicious code.
The cited source says the same thing.
The article says “Aikido names the Packagist package digitalelvis/neuronai-studio 3.14.3, as well as Go module versions github.com/onuragtas/docker-env v1.0.278 and v1.0.280”
The source says “Packagist digitalelvis/neuronai-studio version 3.14.3”
Source cited by the article: aikido.dev/blog/compromised-github-accounts-push-ci-secret-stealer-to-tens-of-thousands-of-repositories
Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN —
Recorded for context only, checked 15 Nov 2026, 20:44 UTC. It played no part in the ranking.