What it does
Security guard plugin for DeepSeek Harness: secret redaction, a shell command denylist, prompt-injection scanning, and syntax verification.
Installation
dsh plugin --profile web add github:okaditya84/dsh-agent-sentinelInstall method: GitHub · not yet tested in container (L3+)
Compatibility
| DSH Version | Status |
|---|---|
| not stated | Declared — not tested |
Requirements
- • Node.js: not stated
- • DSH: declared "not stated"
- • External credentials: none detected
Security Report
Automated static scan, not manual review.
Critical findings in non-blocking categories (dynamic code execution, shell execution, install scripts, obfuscation). Common in CLI/terminal plugins but worth reviewing.
code-exec ×1shell ×10.1.055e5d0af8eca2026-08-1955e5d0af8ecaThe scan result is only valid for the scanned commit. If the latest commit differs, the badge shows outdated until the daily pipeline rescans.
Disclaimer: automated static analysis, not a security guarantee. Always review what you install.
• Static heuristic scan: done (10 files)
• Dependency vulnerabilities: requires deep audit (L3+)
• Permission sandboxing: requires runtime testing (L4+)
High-risk findings · 3 / 3
- criticalChild process module usage (Node.js)src/code-verifier.ts:1import { execFile } from 'node:child_process'
- criticalShell command execution (exec / execSync)src/code-verifier.ts:104const child = execFile(bin, ['-c', 'import sys; compile(sys.stdin.read(), "<dsh-agent-sentinel>", "exec")'], (error, _stdout, stderr) => {
- warningFilesystem write operationssrc/audit-logger.ts:47await fs.appendFile(this.logPath, line, 'utf8')
Heuristic static scan — may produce false positives. Review the source yourself before trusting.
Activity
Last commit 2026-08-18 · activity: Active
• Repo created: 2026-08-19
• Stars: ★ 0 · Forks: 0
• Health: Active — committed within last 30 days