nxtg-ai/Faultline AI Trust & Safety Scanner
Forensic AI output verification — detects hallucination, manipulation, and policy violations in generated content
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| input | File to scan (path relative to repo root, e.g. "docs/ai-output.md") | no | README.md |
| fail-on | Minimum severity to fail the build: critical, high, medium, low | no | high |
| provider | AI provider: gemini, openai, claude, perplexity, mock. Defaults to auto-detect from env vars. | no | "" |
| output-path | Path to write SARIF output | no | faultline.sarif |
Outputs
| name | description |
|---|---|
| sarif-file | Path to generated SARIF file |
| verdict-count | Total number of verdicts returned |
| highest-severity | Highest severity found: high, medium, low, none |