atlas-project-harness@0.1.0-beta.23
践行 Project Harness 的全流程 AI 编程工作流治理引擎
AI Security Review
scanned 5d ago · by lpm-firewall-aiInspected source establishes an automatic, authenticated issue-creation path to the publisher’s default GitHub repository, with no recipient-specific confirmation. This is concrete unrequested account activity and warrants blocking publication.
Decision evidence
public snapshot- The package exposes an Atlas CLI and Pi agent extension.
- For listed design conflicts, its shipped rules direct the agent to submit a report automatically and silently; they treat a user request as a fallback.
- The CLI report command runs the package’s report script and passes through the process environment.
- The reporter defaults to the publisher’s GitHub repository and reads credentials from environment variables or Git’s credential helper.
- The reporter sends the credential in an authenticated request to GitHub’s issue API and creates an issue when a token is available, without a recipient-specific confirmation step.
Source & flagged code
6 flagged · loading sourceSource file is highly similar to a previously finalized malicious package; route for source-aware review.
bin/atlas.jsView on unpkgSource fingerprint signature matches a known malicious package signature; route for source-aware review.
bin/atlas.jsView on unpkgFindings
Affected versions and remediation
This report applies to atlas-project-harness@0.1.0-beta.23.
See all versions of atlas-project-harness for other recorded verdicts.
- Avoid installing atlas-project-harness@0.1.0-beta.23. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.