| name | privacy-guard |
| description | Auto-apply whenever reading, outputting, or sharing file contents that may contain sensitive data, including secrets, API keys, passwords, tokens, credentials, .env files, private keys, certificates, or PII. Enforces strict file-type allowlist and redacts sensitive data. |
Privacy Guard Protocol
File Scope (Strict Allowlist)
ONLY process:
- Code:
.go, .rs, .zig, .ts, .js, .py, .c, .cpp, .h, .css, .html
- Build:
go.mod, go.sum, build.zig*, Cargo.*, package.json, *lock*, requirements.txt, Pipfile, Makefile
- Config:
Dockerfile, *.yaml/yml, .env.example, .gitignore, .editorconfig, .toml, .json (infrastructure, build, tooling, or agent config — e.g., package.json, tsconfig.json, opencode.json), **/skills/**/*.md (skill files)
REJECT immediately:
- Documents:
.pdf, .docx, .doc, .rtf, .pages
- Data:
.xls*, .csv, .numbers, user-record JSON/YAML/XML
- Secrets:
.pem, .key, id_rsa, secrets.*
Privacy Scan (Execute Before Processing)
Detect and redact to <REDACTED>:
- API keys (AWS, Stripe, etc.)
- Database passwords
- Real names (non-author)
- Email addresses (non-dummy)
- Phone numbers
- Physical addresses
- Credit cards
- Internal IPs (
192.168.x.x, 10.x.x.x) → <INTERNAL_IP>
Execution
- Validate file type against allowlist
- Scan for PII/secrets
- Redact matches + report types found
- Proceed with request OR output:
🚫 PRIVACY GUARD: File/content rejected
Never output real PII or use it in examples.
