soulsniper-v2/LintAI Validation
Validates LLM outputs using configurable assertions with confidence scoring
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| prompt | The original prompt sent to the LLM | yes | — |
| output | The LLM output to validate | yes | — |
| assertions-config | Path to assertions configuration file (JSON/YAML) | yes | — |
| pass-threshold | Minimum score threshold for passing (0-100) | no | 70 |
| fail-on-warning | Fail validation if there are warnings | no | false |
Outputs
| name | description |
|---|---|
| passed | Whether validation passed (true/false) |
| score | Validation confidence score (0-100) |
| failed-assertions | Number of failed assertions |
| warnings-count | Number of warnings |