@daniellins/power-claude
Lean, budget-gated agent framework for Claude Code — pods by job-to-be-done
This report shows the evidence for @daniellins/power-claude@0.16.4. 1 of 4 known versions is blocked or flagged. The newest version, 0.17.2, is flagged. See all versions.
AI Security Review
scanned 2mo ago · by lpm-firewall-aiHard-coded reads of unrelated local secret files combined with authenticated outbound requests establish credential-exfiltration behavior. The lack of lifecycle hooks limits automatic activation but does not remove the concrete malicious chain.
Decision evidence
public snapshot- Production executor reads API keys from hard-coded developer home-directory .env paths when environment variables are absent.
- The harvested Anthropic key is sent as an x-api-key to api.anthropic.com.
- The installer distributes the executor as a motor script.
- package.json has no preinstall/install/postinstall lifecycle hook.
- CLI installation is user-invoked; no import-time network activity was found.
Source & flagged code
8 flagged · loading sourcePackage source references child process execution.
scripts/veto-checks/render-gate.cjsView on unpkg · L37Package source references dynamic require/import behavior.
bin/cli.jsView on unpkg · L17Findings
Affected versions and remediation
This report applies to @daniellins/power-claude@0.16.4.
The versions list below shows the verdict for each recorded version.
- Avoid installing @daniellins/power-claude@0.16.4. 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.