@papi-ai/server@0.7.105
PAPI MCP server — AI-powered sprint planning, build execution, and strategy review for software projects
AI Security Review
scanned 1mo ago · by lpm-firewall-aiSource supports a behavior-specific warning for explicit agent configuration mutation and persistent executable integration. No concrete malicious chain was established, so publication blocking is unwarranted.
Decision evidence
public snapshot- Explicit project setup adds persistent permission allowances for all PAPI tools to Claude settings.
- Local setup installs a frontend agent, skill, and executable hook when a frontend stack is detected.
- The statusline installer requires --install and registers an unpinned PAPI CLI command in user-level Claude settings.
- package.json has no registry-install lifecycle hooks or runtime self-dependency.
- The flagged backfill location runs Git history queries for documentation paths rather than modifying agent configuration.
- The configuration changes belong to explicit PAPI setup operations; no unconsented npm install-time takeover was identified.
- The statusline installer preserves an existing statusline, and its authentication uses credentials supplied for the configured PAPI endpoint.
- CLI setup obtains a PAPI credential through browser-approved device authorization rather than harvesting an unrelated service credential.
Source & flagged code
6 flaggedPackage source references child process execution.
dist/backfill-cycle-metrics.jsView on unpkg · L266Manifest-reachable source sends a prompted API credential to a fixed unofficial gateway and persists the redirection.
Findings
Affected versions and remediation
This report applies to @papi-ai/server@0.7.105.
See all versions of @papi-ai/server for other recorded verdicts.
- Review the evidence and your use of @papi-ai/server@0.7.105 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.