@saluzi/saluzi-edu@0.7.3
Saluzi CLI - interactive AI coding assistant in the terminal
AI security review
scanned · by lpm-firewall-aiThe package automatically writes to a global AI assistant settings file during postinstall, changing permissions without an explicit user request. The added rules are protective, but the install-time modification still matches the policy for AI-agent control-surface changes.
Decision evidence
public snapshot- The postinstall lifecycle hook runs scripts/install-defaults.mjs.
- The script resolves a global settings file at ~/.saluzi-edu/settings.json, or settings.json under SALUZI_CONFIG_DIR.
- It adds permission-deny rules for commands including rm -rf, dropping databases, force-pushing, fork bombs, and mkfs.
- The hook writes the resulting settings file automatically during installation.
- The added rules restrict destructive commands and are merged into the existing settings.
- The script skips writing when no settings changes are needed.
- 127.0.0.1
- about.gitlab.com
- about.meta.com
- about.x.com
- airtable.com
- angular.dev
- apache.org
- api.anthropic.com
- asana.com
- atlassian.design
- blog.golang.org
- brand.databricks.com
- brand.zoom.us
- brandland.zendesk.com
- business.pinterest.com
- cdn.jsdelivr.net
- chat.mistral.ai
- chevrotain.io
- circleci.com
- claude.ai
- cloud.google.com
- cloudinary.com
- discord.com
- docs.anthropic.com
- docs.expo.dev
- docs.oasis-open.org
- en.wikipedia.org
- example.com
- firebase.google.com
- fly.io
- gemini.google.com
- get.jenkins.io
- github.com
- grafana.com
- helm.sh
- huggingface.co
- influxdata.github.io
- json-schema.org
- langium.org
- linear.app
- mariadb.com
- miro.com
- neon.com
- newsroom.paypal-corp.com
- nodejs.org
- openai.com
- openoffice.org
- openrouter.ai
- opensearch.org
- planetscale.com
Findings
3 high · 8 medium · 7 lowWhere they are, most severe file first. Code opens a file's flagged lines, with their evidence, below.
package.json1 highdist/guide/assets/chunk-727SXJPM-C91w5wT3.js#virtual:normalized:round11 highdist/guide/assets/diagram-OG6HWLK6-B5R3Z1Zi.js1 highdist/rcs-web/assets/motion-BILbV7YQ.js1 mediumdist/vendor/audio-capture/x64-darwin/audio-capture.node1 mediumdist/tree-sitter.wasm1 mediumdist/rcs-web/assets/shiki-ChKhlm5h.js1 mediumdist/cli.js1 mediumShow all 18 findings
package.json1 findingdist/guide/assets/chunk-727SXJPM-C91w5wT3.js#virtual:normalized:round11 findingdist/guide/assets/diagram-OG6HWLK6-B5R3Z1Zi.js1 findingdist/rcs-web/assets/motion-BILbV7YQ.js1 findingdist/vendor/audio-capture/x64-darwin/audio-capture.node1 findingdist/tree-sitter.wasm1 findingdist/rcs-web/assets/shiki-ChKhlm5h.js1 findingdist/cli.js1 findingSource & flagged code
8 flaggedPackage defines install-time lifecycle scripts.
package.jsonView on unpkgScanner record · 1 line
Package source references dynamic require/import behavior.
dist/rcs-web/assets/motion-BILbV7YQ.jsView on unpkg · L1Code excerpt · 1 line
Package ships native binary artifacts.
dist/vendor/audio-capture/x64-darwin/audio-capture.nodeView on unpkgScanner record · 4 lines
Scanner record · 4 lines
Package contains source files above the normal full-analysis size ceiling.
dist/rcs-web/assets/shiki-ChKhlm5h.jsView on unpkgScanner record · 4 lines
Package contains an oversized executable-looking CLI entrypoint.
dist/cli.jsView on unpkgScanner record · 4 lines
A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/guide/assets/chunk-727SXJPM-C91w5wT3.js#virtual:normalized:round1View on unpkgScanner record · 1 line
Source file is highly similar to a previously finalized malicious package; route for source-aware review.
dist/guide/assets/diagram-OG6HWLK6-B5R3Z1Zi.jsView on unpkgScanner record · 7 lines
Affected versions and remediation
This report applies to @saluzi/saluzi-edu@0.7.3.
See all versions of @saluzi/saluzi-edu for other recorded verdicts.
- Avoid installing @saluzi/saluzi-edu@0.7.3. 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, check your AI agents' configuration and instruction files (the files this report lists) for changes you did not make, and remove them. This finding does not by itself establish malware intent.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.