·
Last changedVersion:
This version did not pass the hardening gate. Hardening updated the skill content, deleted the lines high-severity poisoning findings flagged in 1 file before any model saw the skill, and independently checked the resulting changes. Classified findings before hardening: 10. Classified findings closed: 0. Classified findings remaining: 10. Unclassified scanner findings remaining: 0. Review findings closed: 17. Review findings remaining: 4.
2
7
1
Expand | File | Group | Subtype | Severity | Status |
|---|---|---|---|---|---|
SKILL.md | Obfuscation & Evasion | Trigger-based activation | High | Open | |
references/video-summarization-environment-variables.md | Prompt Injection | — | High | Open | |
SKILL.md | Sensitive Data & Secret Exposure | Hardcoded secret | Medium | Open | |
references/end-to-end-example.md | Sensitive Data & Secret Exposure | Data exfiltration | Medium | Open | |
references/hitl-prompts.md | Sensitive Data & Secret Exposure | Data exfiltration | Medium | Open | |
references/video-summarization-api.md | Sensitive Data & Secret Exposure | Data exfiltration | Medium | Open | |
BENCHMARK.md | Sensitive Data & Secret Exposure | Agent snooping | Medium | Open | |
evals/lvs_api_ops.json | Sensitive Data & Secret Exposure | Data exfiltration | Medium | Open |
10 findings
No skill file found.
Bring Chainguard's hardening pipeline to your first-party skills — custom rulesets, automated remediation, full audit trail.
skills.cgr.dev/public/github.com/nvidia/skills/vss-summarize-video:latest
Apache-2.0
HARDENING.md