{"items":[{"id":"NVI-00012039","kind":"vulnerability","title":"Pydantic AI: Event loop blocked by quadratic title extraction in `web_fetch`","revision":4,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected.","NVD lastModified does not establish freshness of every associated CPE record."],"lifecycle":"active","identifiers":["GHSA-fpf4-vwcp-v4hp","CVE-2026-107290"],"plain_summary":"GitHub lists this reviewed advisory for pydantic-ai, pydantic-ai, pydantic-ai-slim, pydantic-ai-slim. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-08b57fdb6e50e05d87d0eadf64b23e55"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-08T16:48:27.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-9204f5c2e67a26fbcd44a49accce790c"},{"kind":"CVSS","as_of":"2026-10-13T14:34:38.623Z","value":"6.5 MEDIUM (3.1)","meaning":"security-advisories@github.com severity; no combined Null risk score.","source_id":"sr-4a50f78bbe6c591a2b6b85b168972b9d"},{"kind":"EPSS","as_of":"2026-11-05T12:03:04.000Z","value":"0.512%","meaning":"FIRST EPSS model v2026.06.15; exploitation probability over 30 days, percentile 0.43814. Missing scores are unknown.","source_id":"sr-08b57fdb6e50e05d87d0eadf64b23e55"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-9204f5c2e67a26fbcd44a49accce790c","url":"https://github.com/advisories/GHSA-fpf4-vwcp-v4hp","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-fpf4-vwcp-v4hp","publisher":"GitHub Advisory Database","observed_at":"2026-11-01T00:58:38.867Z"},{"id":"sr-4a50f78bbe6c591a2b6b85b168972b9d","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-107290","basis":"official_catalog","notes":"This product uses data from the NVD API but is not endorsed or certified by the NVD.","title":"CVE-2026-107290","publisher":"NIST National Vulnerability Database","observed_at":"2026-11-03T09:58:29.490Z"},{"id":"sr-08b57fdb6e50e05d87d0eadf64b23e55","url":"https://www.first.org/epss/data.html","basis":"official_catalog","notes":"FIRST EPSS; probability of exploitation in the next 30 days, not local compromise. Dataset date/model retained; FIRST services terms apply.","title":"CVE-2026-107290","publisher":"FIRST EPSS","observed_at":"2026-11-06T00:24:57.626Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 1.77.0, < 1.107.6"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 2.0.0b1, < 2.44.0"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai-slim","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 1.77.0, < 1.107.6"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai-slim","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 2.0.0b1, < 2.44.0"]}],"null_published_at":"2026-11-08T15:11:49.411Z","source_updated_at":"2026-10-08T16:48:27.000Z","source_disclosed_at":"2026-10-08T16:48:25.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"Dated EPSS estimate updated; this is not a new disclosure or proof of exploitation.","null_first_observed_at":"2026-11-01T00:58:38.867Z"},{"id":"NVI-00012039","kind":"vulnerability","title":"Pydantic AI: Event loop blocked by quadratic title extraction in `web_fetch`","revision":3,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected.","NVD lastModified does not establish freshness of every associated CPE record."],"lifecycle":"active","identifiers":["GHSA-fpf4-vwcp-v4hp","CVE-2026-107290"],"plain_summary":"GitHub lists this reviewed advisory for pydantic-ai, pydantic-ai, pydantic-ai-slim, pydantic-ai-slim. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-6fcb15bc305f3222e24b6aa3f5d6e8f7"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-08T16:48:27.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-9204f5c2e67a26fbcd44a49accce790c"},{"kind":"CVSS","as_of":"2026-10-13T14:34:38.623Z","value":"6.5 MEDIUM (3.1)","meaning":"security-advisories@github.com severity; no combined Null risk score.","source_id":"sr-4a50f78bbe6c591a2b6b85b168972b9d"},{"kind":"EPSS","as_of":"2026-10-31T12:00:27.000Z","value":"0.512%","meaning":"FIRST EPSS model v2026.06.15; exploitation probability over 30 days, percentile 0.43676. Missing scores are unknown.","source_id":"sr-6fcb15bc305f3222e24b6aa3f5d6e8f7"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-9204f5c2e67a26fbcd44a49accce790c","url":"https://github.com/advisories/GHSA-fpf4-vwcp-v4hp","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-fpf4-vwcp-v4hp","publisher":"GitHub Advisory Database","observed_at":"2026-11-01T00:58:38.867Z"},{"id":"sr-4a50f78bbe6c591a2b6b85b168972b9d","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-107290","basis":"official_catalog","notes":"This product uses data from the NVD API but is not endorsed or certified by the NVD.","title":"CVE-2026-107290","publisher":"NIST National Vulnerability Database","observed_at":"2026-11-03T09:58:29.490Z"},{"id":"sr-6fcb15bc305f3222e24b6aa3f5d6e8f7","url":"https://www.first.org/epss/data.html","basis":"official_catalog","notes":"FIRST EPSS; probability of exploitation in the next 30 days, not local compromise. Dataset date/model retained; FIRST services terms apply.","title":"CVE-2026-107290","publisher":"FIRST EPSS","observed_at":"2026-11-01T01:39:43.065Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 1.77.0, < 1.107.6"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 2.0.0b1, < 2.44.0"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai-slim","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 1.77.0, < 1.107.6"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai-slim","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 2.0.0b1, < 2.44.0"]}],"null_published_at":"2026-11-05T19:10:23.248Z","source_updated_at":"2026-10-08T16:48:27.000Z","source_disclosed_at":"2026-10-08T16:48:25.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"Source facts changed; compare the preserved prior revision.","null_first_observed_at":"2026-11-01T00:58:38.867Z"},{"id":"NVI-00012039","kind":"vulnerability","title":"Pydantic AI: Event loop blocked by quadratic title extraction in `web_fetch`","revision":2,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected."],"lifecycle":"active","identifiers":["GHSA-fpf4-vwcp-v4hp","CVE-2026-107290"],"plain_summary":"GitHub lists this reviewed advisory for pydantic-ai, pydantic-ai, pydantic-ai-slim, pydantic-ai-slim. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-6fcb15bc305f3222e24b6aa3f5d6e8f7"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-08T16:48:27.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-9204f5c2e67a26fbcd44a49accce790c"},{"kind":"EPSS","as_of":"2026-10-31T12:00:27.000Z","value":"0.512%","meaning":"FIRST EPSS model v2026.06.15; exploitation probability over 30 days, percentile 0.43676. Missing scores are unknown.","source_id":"sr-6fcb15bc305f3222e24b6aa3f5d6e8f7"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-9204f5c2e67a26fbcd44a49accce790c","url":"https://github.com/advisories/GHSA-fpf4-vwcp-v4hp","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-fpf4-vwcp-v4hp","publisher":"GitHub Advisory Database","observed_at":"2026-11-01T00:58:38.867Z"},{"id":"sr-6fcb15bc305f3222e24b6aa3f5d6e8f7","url":"https://www.first.org/epss/data.html","basis":"official_catalog","notes":"FIRST EPSS; probability of exploitation in the next 30 days, not local compromise. Dataset date/model retained; FIRST services terms apply.","title":"CVE-2026-107290","publisher":"FIRST EPSS","observed_at":"2026-11-01T01:39:43.065Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 1.77.0, < 1.107.6"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 2.0.0b1, < 2.44.0"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai-slim","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 1.77.0, < 1.107.6"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai-slim","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 2.0.0b1, < 2.44.0"]}],"null_published_at":"2026-11-03T10:03:21.788Z","source_updated_at":"2026-10-08T16:48:27.000Z","source_disclosed_at":"2026-10-08T16:48:25.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"Dated EPSS estimate updated; this is not a new disclosure or proof of exploitation.","null_first_observed_at":"2026-11-01T00:58:38.867Z"},{"id":"NVI-00012039","kind":"vulnerability","title":"Pydantic AI: Event loop blocked by quadratic title extraction in `web_fetch`","revision":1,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected."],"lifecycle":"active","identifiers":["GHSA-fpf4-vwcp-v4hp","CVE-2026-107290"],"plain_summary":"GitHub lists this reviewed advisory for pydantic-ai, pydantic-ai, pydantic-ai-slim, pydantic-ai-slim. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-08T16:48:27.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-9204f5c2e67a26fbcd44a49accce790c"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-9204f5c2e67a26fbcd44a49accce790c","url":"https://github.com/advisories/GHSA-fpf4-vwcp-v4hp","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-fpf4-vwcp-v4hp","publisher":"GitHub Advisory Database","observed_at":"2026-11-01T00:58:38.867Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 1.77.0, < 1.107.6"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 2.0.0b1, < 2.44.0"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai-slim","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 1.77.0, < 1.107.6"]},{"certainty":"source_listed","source_ids":["sr-9204f5c2e67a26fbcd44a49accce790c"],"technology":"pydantic-ai-slim","known_exclusions":[],"conditions_summary":"### Summary\n\nThe local web-fetch tool (`web_fetch_tool`, also used as the `WebFetch` capability's local fallback) processed responses with several steps whose running time grows quadratically with the size of certain server-controlled inputs, and ran them on the event loop: decoding the body with whichever charset the server declared, extracting the page title with a backtracking regular expression, and converting the HTML to markdown. An application that exposes this tool to untrusted prompts can be steered to fetch an attacker-controlled page of a megabyte or two that blocks the event loop for minutes, stalling every other coroutine in the process — other agent runs, other requests being served — for the duration.\n\nThis is an **availability** issue only. SSRF protections and the download size limit introduced in GHSA-v2xh-2vp8-57h8 are unaffected; that limit bounds how much is downloaded, not how long the response takes to process.\n\n### Details\n\nTitle extraction used a backtracking pattern over the raw response body, so a body made of repeated unterminated tag openings cost time proportional to the square of its size. The HTML-to-markdown conversion had the same shape in three of","public_affected_ranges":[">= 2.0.0b1, < 2.44.0"]}],"null_published_at":"2026-11-01T01:03:56.624Z","source_updated_at":"2026-10-08T16:48:27.000Z","source_disclosed_at":"2026-10-08T16:48:25.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"First indexed by Null; source disclosure date is shown separately.","null_first_observed_at":"2026-11-01T00:58:38.867Z"}],"next_cursor":null,"snapshot_revision":"42594"}