·
Last changedVersion:
This version did not pass the hardening gate. Hardening applied 15 targeted changes, deleted the lines high-severity poisoning findings flagged in 5 files before any model saw the skill, and preserved all other text. Classified findings before hardening: 29. Classified findings closed: 1. Classified findings remaining: 28. Unclassified scanner findings remaining: 3. Review findings closed: 10. Review findings remaining: 1.
10
9
10
Expand | File | Group | Subtype | Severity | Status |
|---|---|---|---|---|---|
references/servers-infrastructure.md | Privilege Escalation & Host / Cloud Compromise | Privilege escalation | High | Open | |
references/http-networking.md | Tool & MCP Manipulation | Unauthorized tool use | High | Open | |
references/html-markup.md | Prompt Injection | Direct prompt injection | High | Open | |
references/media-graphics.md | Prompt Injection | Direct prompt injection | High | Open | |
references/performance-optimization.md | Prompt Injection | Direct prompt injection | High | Open | |
references/development-tools.md | Privilege Escalation & Host / Cloud Compromise | Privilege escalation | High | Open | |
references/architecture-patterns.md | Tool & MCP Manipulation | Unauthorized tool use | High | Open | |
references/security-authentication.md | Privilege Escalation & Host / Cloud Compromise | Privilege escalation | High | Open |
29 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/github/skills/web-coder:latest
MIT
HARDENING.md