Weekly CSV feeds of semi-static open proxy exit IPs (IPv4 + IPv6), plus high-activity subnet extracts.
Intended for defensive workflows: abuse mitigation, DNSBL/RBL tuning, monitoring, and research.
Companion dataset to Open Proxy Life Expectancy (2022–2024).
"Semi-static" = proxies detectable long enough to be useful as a stability-filtered snapshot, vs. transient one-off detections.
| Term | Meaning |
|---|---|
| Exit IP | Source address observed when a proxy relays traffic |
| Entry IP | Address of the proxy service contacted by the scanner |
| Asymmetric proxy | Entry and exit IP differ; entry IP recorded in comment column |
Entry ports and protocols are omitted to reduce misuse risk.
| File | Description |
|---|---|
proxy_exits_7d_ipv4.csv |
IPv4 exits active ≥7 days |
proxy_exits_7d_ipv6.csv |
IPv6 exits active ≥7 days |
proxy_exits_30d_ipv4.csv |
IPv4 exits active ≥30 days |
proxy_exits_30d_ipv6.csv |
IPv6 exits active ≥30 days |
Subnets with ≥16 unique exit IPs:
| File | Description |
|---|---|
proxy_subnets_7d_ipv4.csv |
IPv4 /24s active ≥7 days |
proxy_subnets_7d_ipv6.csv |
IPv6 subnets active ≥7 days |
proxy_subnets_30d_ipv4.csv |
IPv4 /24s active ≥30 days |
proxy_subnets_30d_ipv6.csv |
IPv6 subnets active ≥30 days |
Methodology:
- IPv4: /24 bucketing
- IPv6: fitted dynamic prefix (/4 boundary) to capture dense exit ranges; limited to exits tied to specific entry IPs
Semi-static exit counts for ≥7d and ≥30d cohorts.
High-activity subnets (≥16 exits) by cohort.
IPv4 vs IPv6 composition.
Defensive security research, measurement, monitoring, and abuse mitigation.
Not for circumventing safeguards or enabling malicious activity. If you believe something here creates unintended harm, please contact me.
This data is provided as-is, without warranty. Use freely; attribution appreciated but not required.