·
Last changedVersion:
This version did not pass the hardening gate. Hardening applied 15 targeted changes, deleted the lines high-severity poisoning findings flagged in 1 file before any model saw the skill, and preserved all other text. Classified findings before hardening: 8. Classified findings closed: 2. Classified findings remaining: 6. Unclassified scanner findings remaining: 0. Review findings closed: 18. Review findings remaining: 4.
2
3
3
Expand | File | Group | Subtype | Severity | Status |
|---|---|---|---|---|---|
scripts/skill_registry_ops.py | Sensitive Data & Secret Exposure | — | High | Open | |
references/manage-skills.md | Excessive Agency & Rogue Autonomy | — | High | Open | |
scripts/skill_registry_ops.py | Sensitive Data & Secret Exposure | Data exfiltration | Medium | Open | |
SKILL.md | Tool & MCP Manipulation | MCP over-permissioning | Medium | Open | |
SKILL.md | Excessive Agency & Rogue Autonomy | — | Medium | Open | |
SKILL.md | Malicious Code & Command Execution | — | Low | Closed | |
SKILL.md | Supply Chain & Install-Time Trust | Install-time download and execute | Low | Closed | |
scripts/requirements.txt | Supply Chain & Install-Time Trust | Dependency or plugin compromise | Low | Open |
8 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/google/skills/cloud/agent-platform-skill-registry:latest
Apache-2.0
HARDENING.md