blackvectorops/Semantic Firewall
Behavioral analysis engine for Go. Detects malware, proves loop equivalence, and enforces semantic immutability.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Activelast commit Jul 4, 2026
- License
- MIT
Pinned Snippet
uses: blackvectorops/semantic_firewall@de79499c295c2b3947439f11bab85826f1060275 # v4.2.0tags can be moved; commit SHAs can't. why a SHA?
Inputs
| name | description | required | default |
|---|---|---|---|
| path | Path to file or directory to analyze | no | . |
| mode | Operation mode: check, diff, scan, or audit | no | check |
| go-version | Go version to use | no | 1.24 |
| signatures | Path to signature database (scan mode only). Required for scan mode. | no | "" |
| threshold | Match confidence threshold 0.0-1.0 (scan mode only) | no | 0.75 |
| strict | Enable strict mode validation (check mode only) | no | false |
| fail-on-alert | Fail workflow if malware detected (scan mode only) | no | true |
| scan-deps | Scan imported dependencies (scan mode only) | no | false |
| deps-depth | Dependency scan depth: direct or transitive (scan mode only) | no | direct |
| api-key | LLM API Key (audit mode only). Required for intent verification. | no | — |
| model | LLM Model to use (audit mode only). | no | gpt-4 |
| api-base | Custom LLM API Base URL (audit mode only). | no | — |
Outputs
| name | description |
|---|---|
| result | JSON output from the selected mode |
| alert-count | Number of malware alerts (scan mode) |
| deps-scanned | Number of dependency functions scanned (scan mode with --deps) |
| semantic-match-pct | Semantic match percentage (diff mode) |
| audit-verdict | Final audit verdict (audit mode): PASS or FAIL |