### Summary The Docker API server's SSRF protection (`validate_webhook_url` / `validate_url_destination` in `deploy/docker/utils.py`) used an explicit IPv4/IPv6 CIDR blocklist that missed several address families. An attacker could reach internal services and cloud metadata endpoints (e.g. `169.254.169.254`) despite the filter by encoding an internal IPv4 address inside an IPv6 transition form, or by using the IPv6 unspecified address. Because the Docker API is unauthenticated by default (`jwt_enabled: false`), no credentials are required. ### Affected paths The blocklist was applied to crawl URLs (`POST /crawl`, `/md`, `/html`, `/screenshot`, `/pdf`, `/execute_js`) and webhook URLs (`/crawl/job`, `/llm/job`). All shared the same incomplete check. ### Bypasses The following all resolve to (or route to) blocked internal addresses but were NOT caught: - IPv6 unspecified `::` - NAT64 `64:ff9b::a9fe:a9fe` (embeds `169.254.169.254`) - 6to4 `2002:a9fe:a9fe::` (embeds `169.254.169.254`) - IPv4-mapped `::ffff:169.254.169.254` - IPv4-compatible `::a9fe:a9fe` The error message also echoed the resolved internal IP, acting as a minor DNS/oracle leak. ### Impact Server-Side Request Forgery: an unauthenticated attacker can make the server fetch internal-network URLs and cloud instance-metadata endpoints, potentially exposing internal services and cloud credentials. ### Fix The blocklist is replaced by a single rule: reject any resolved IP where `not ip.is_global`, evaluated on the address AND every embedded IPv4 transition form (v4-mapped, NAT64 `64:ff9b::/96`, 6to4 `2002::/16`, v4-compat `::/96`). Error messages are now opaque and no longer echo the resolved IP. ### Workarounds - Upgrade to the patched version. - Enable authentication (`CRAWL4AI_API_TOKEN`). - Restrict the container's outbound network access (egress firewall / no metadata route). ### Credits Internal security audit (Crawl4AI maintainers).
### Summary The Docker API server's SSRF protection (`validate_webhook_url` / `validate_url_destination` in `deploy/docker/utils.py`) used an explicit IPv4/IPv6 CIDR blocklist that missed several address families. An attacker could reach internal services and cloud metadata endpoints (e.g. `169.254.169.254`) despite the filter by encoding an internal IPv4 address inside an IPv6 transition form, or by using the IPv6 unspecified address. Because the Docker API is unauthenticated by default (`jwt_enabled: false`), no credentials are required. ### Affected paths The blocklist was applied to crawl URLs (`POST /crawl`, `/md`, `/html`, `/screenshot`, `/pdf`, `/execute_js`) and webhook URLs (`/crawl/job`, `/llm/job`). All shared the same incomplete check. ### Bypasses The following all resolve to (or route to) blocked internal addresses but were NOT caught: - IPv6 unspecified `::` - NAT64 `64:ff9b::a9fe:a9fe` (embeds `169.254.169.254`) - 6to4 `2002:a9fe:a9fe::` (embeds `169.254.169.254`) - IPv4-mapped `::ffff:169.254.169.254` - IPv4-compatible `::a9fe:a9fe` The error message also echoed the resolved internal IP, acting as a minor DNS/oracle leak. ### Impact Server-Side Request Forgery: an unauthenticated attacker can make the server fetch internal-network URLs and cloud instance-metadata endpoints, potentially exposing internal services and cloud credentials. ### Fix The blocklist is replaced by a single rule: reject any resolved IP where `not ip.is_global`, evaluated on the address AND every embedded IPv4 transition form (v4-mapped, NAT64 `64:ff9b::/96`, 6to4 `2002::/16`, v4-compat `::/96`). Error messages are now opaque and no longer echo the resolved IP. ### Workarounds - Upgrade to the patched version. - Enable authentication (`CRAWL4AI_API_TOKEN`). - Restrict the container's outbound network access (egress firewall / no metadata route). ### Credits Internal security audit (Crawl4AI maintainers).
Update crawl4ai to 0.8.8 if you use the affected versions. Test the change in a non-production environment first.
Local check
hol-guard supply-chain scanCrawl4AI: SSRF filter bypass in Docker server via IPv6 transition forms (NAT64 / 6to4 / unspecified / v4-mapped) affects crawl4ai (pip). Severity is high. ### Summary The Docker API server's SSRF protection (`validate_webhook_url` / `validate_url_destination` in `deploy/docker/utils.py`) used an explicit IPv4/IPv6 CIDR blocklist that missed several address families. An attacker could reach internal services and cloud metadata endpoints (e.g. `169.254.169.254`) despite the filter by encoding an internal IPv4 address inside an IPv6 transition form, or by using the IPv6 unspecified address. Because the Docker API is unauthenticated by default (`jwt_enabled: false`), no credentials are required. ### Affected paths The blocklist was applied to crawl URLs (`POST /crawl`, `/md`, `/html`, `/screenshot`, `/pdf`, `/execute_js`) and webhook URLs (`/crawl/job`, `/llm/job`). All shared the same incomplete check. ### Bypasses The following all resolve to (or route to) blocked internal addresses but were NOT caught: - IPv6 unspecified `::` - NAT64 `64:ff9b::a9fe:a9fe` (embeds `169.254.169.254`) - 6to4 `2002:a9fe:a9fe::` (embeds `169.254.169.254`) - IPv4-mapped `::ffff:169.254.169.254` - IPv4-compatible `::a9fe:a9fe` The error message also echoed the resolved internal IP, acting as a minor DNS/oracle leak. ### Impact Server-Side Request Forgery: an unauthenticated attacker can make the server fetch internal-network URLs and cloud instance-metadata endpoints, potentially exposing internal services and cloud credentials. ### Fix The blocklist is replaced by a single rule: reject any resolved IP where `not ip.is_global`, evaluated on the address AND every embedded IPv4 transition form (v4-mapped, NAT64 `64:ff9b::/96`, 6to4 `2002::/16`, v4-compat `::/96`). Error messages are now opaque and no longer echo the resolved IP. ### Workarounds - Upgrade to the patched version. - Enable authentication (`CRAWL4AI_API_TOKEN`). - Restrict the container's outbound network access (egress firewall / no metadata route). ### Credits Internal security audit (Crawl4AI maintainers).
AI coding agents often install or upgrade packages automatically in pip. A high vulnerability in a dependency can be pulled into a project through a normal install or update without a human reviewing the change, expanding the blast radius from a single package to every agent workspace that depends on it.
| Package | Affected range | Fixed version |
|---|---|---|
| crawl4aipip | <=0.8.7 | 0.8.8 |
Fixed versions are reported by the source feed; confirm compatibility before updating.
Reported by GitHub Security Advisories (ghsa).
HOL Guard can help your team review package activity against supported protection paths.
Explore HOL GuardUpdate crawl4ai to 0.8.8 if you use the affected versions. Test the change in a non-production environment first.
Local check
hol-guard supply-chain scanCrawl4AI: SSRF filter bypass in Docker server via IPv6 transition forms (NAT64 / 6to4 / unspecified / v4-mapped) affects crawl4ai (pip). Severity is high. ### Summary The Docker API server's SSRF protection (`validate_webhook_url` / `validate_url_destination` in `deploy/docker/utils.py`) used an explicit IPv4/IPv6 CIDR blocklist that missed several address families. An attacker could reach internal services and cloud metadata endpoints (e.g. `169.254.169.254`) despite the filter by encoding an internal IPv4 address inside an IPv6 transition form, or by using the IPv6 unspecified address. Because the Docker API is unauthenticated by default (`jwt_enabled: false`), no credentials are required. ### Affected paths The blocklist was applied to crawl URLs (`POST /crawl`, `/md`, `/html`, `/screenshot`, `/pdf`, `/execute_js`) and webhook URLs (`/crawl/job`, `/llm/job`). All shared the same incomplete check. ### Bypasses The following all resolve to (or route to) blocked internal addresses but were NOT caught: - IPv6 unspecified `::` - NAT64 `64:ff9b::a9fe:a9fe` (embeds `169.254.169.254`) - 6to4 `2002:a9fe:a9fe::` (embeds `169.254.169.254`) - IPv4-mapped `::ffff:169.254.169.254` - IPv4-compatible `::a9fe:a9fe` The error message also echoed the resolved internal IP, acting as a minor DNS/oracle leak. ### Impact Server-Side Request Forgery: an unauthenticated attacker can make the server fetch internal-network URLs and cloud instance-metadata endpoints, potentially exposing internal services and cloud credentials. ### Fix The blocklist is replaced by a single rule: reject any resolved IP where `not ip.is_global`, evaluated on the address AND every embedded IPv4 transition form (v4-mapped, NAT64 `64:ff9b::/96`, 6to4 `2002::/16`, v4-compat `::/96`). Error messages are now opaque and no longer echo the resolved IP. ### Workarounds - Upgrade to the patched version. - Enable authentication (`CRAWL4AI_API_TOKEN`). - Restrict the container's outbound network access (egress firewall / no metadata route). ### Credits Internal security audit (Crawl4AI maintainers).
AI coding agents often install or upgrade packages automatically in pip. A high vulnerability in a dependency can be pulled into a project through a normal install or update without a human reviewing the change, expanding the blast radius from a single package to every agent workspace that depends on it.
| Package | Affected range | Fixed version |
|---|---|---|
| crawl4aipip | <=0.8.7 | 0.8.8 |
Fixed versions are reported by the source feed; confirm compatibility before updating.
Reported by GitHub Security Advisories (ghsa).
HOL Guard can help your team review package activity against supported protection paths.
Explore HOL Guard