bad-antics/Prompt Armor — AI Injection Scanner
Scan prompt files and templates for AI injection attacks. 8-layer detection, 17 attack vectors.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| paths | Glob pattern for files to scan (default: prompts/, templates/, *.txt, *.md) | no | . |
| threshold | Minimum threat level to fail: suspicious, hostile, or critical (default: hostile) | no | hostile |
| output-format | Output format: text or json | no | text |
| max-files | Maximum number of files to scan | no | 100 |
Outputs
| name | description |
|---|---|
| total-scanned | Total files scanned |
| threats-found | Number of threats above threshold |
| highest-threat | Highest threat level found |
| report | JSON scan report |