{"answers":{"affected":"5 open source packages ship an affected version, including langsmith.","exploited":"There is no known exploitation of CVE-2026-41182 in the wild and no public exploit code.","fix":"Upgrade langsmith to a fixed version for your release line: 0.7.31 or 0.5.19.","what":"CVE-2026-41182 is a medium-severity vulnerability in langchain-ai langsmith-sdk, published on 23 Apr 2026. An attacker can exploit it over the internet without logging in, with no action from the victim."},"fix_prompt":"Check this codebase for CVE-2026-41182 and fix it if it is affected.\n\nWhat it is: CVE-2026-41182 is a medium-severity vulnerability in langchain-ai langsmith-sdk, published on 23 Apr 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.\nExploitation: There is no known exploitation of CVE-2026-41182 in the wild and no public exploit code.\n\nAffected packages and fixed versions:\n- langsmith (PyPI): affected all versions; fixed in 0.7.31\n- langsmith (PyPI): affected all versions; fixed in 0.7.31\n- langsmith (PyPI): affected all versions; fixed in 0.7.31\n- langsmith (npm): affected all versions; fixed in 0.5.19\n- langsmith (npm): affected all versions; fixed in 0.5.19\n\nTask:\n1. Search the dependency manifests and lockfiles (for example package.json, package-lock.json, requirements.txt, poetry.lock, uv.lock, pom.xml, build.gradle, go.mod, Cargo.lock, Gemfile.lock) and any Dockerfiles for the affected packages, including transitive dependencies.\n2. If an affected version is in use, upgrade it to the fixed version for its release line with the project's package manager, so the lockfile is regenerated. Prefer the smallest upgrade that fixes it.\n3. If there is no fix, or the upgrade would break the build, do not guess: explain the risk and the safest workaround instead.\n4. Run the tests and summarise what changed and why.\n5. If nothing in the project is affected, say so and change nothing.\n\nReference: https://insights.sbomify.com/CVE-2026-41182/ (Markdown: https://insights.sbomify.com/CVE-2026-41182/index.md)","id":"CVE-2026-41182","markdown":"https://insights.sbomify.com/CVE-2026-41182/index.md","name":"CVE-2026-41182","record":{"aliases":["ECHO-e53b-c949-a5db","GHSA-rr7j-v2q5-chgv","MINI-29wm-j3h3-g45r","MINI-3xr3-4qc3-f826","MINI-4cgg-hp2p-4q9j","MINI-5g2c-v2pv-g4cp","MINI-67xj-qwgr-j338","MINI-7hq3-5h9w-px76","MINI-84c4-qcwq-6cc7","MINI-956c-78cg-fv42","MINI-f5vh-mc63-cjhg","MINI-qm38-f8pp-rxqv","MINI-xc79-q2jv-5mf7","PYSEC-2026-2583","ROOT-APP-NPM-CVE-2026-41182","ROOT-APP-NPM-GHSA-rr7j-v2q5-chgv","ROOT-APP-PYPI-CVE-2026-41182","ROOT-APP-PYPI-GHSA-rr7j-v2q5-chgv"],"analysis":{"component_type":"library","fix_type":"upgrade","impacts":["data_read"],"network_reachable":"yes","non_default_config":false,"product_category":"LangSmith SDK","requires_auth":"none","summary":"The LangSmith SDK fails to redact sensitive data in streaming token events, allowing attackers or unauthorized users to view raw LLM outputs.","user_interaction":"not_required","vuln_class":"information_disclosure"},"analysis_generated_by":"qwen3.8-flash-next","assigner":"GitHub_M","cpe":[],"cvss":{"score":5.3,"severity":"medium","source":"cna","vector":"CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N","version":"3.1"},"cwes":[{"id":"CWE-532","name":"Insertion of Sensitive Information into Log File"},{"id":"CWE-359","name":"Exposure of Private Personal Information to an Unauthorized Actor"},{"id":"CWE-200","name":"Exposure of Sensitive Information to an Unauthorized Actor"}],"description":"LangSmith Client SDKs provide SDK's for interacting with the LangSmith platform. Prior to version 0.5.19 of the JavaScript SDK and version 0.7.31 of the Python SDK, the LangSmith SDK's output redaction controls (hideOutputs in JS, hide_outputs in Python) do not apply to streaming token events. When an LLM run produces streaming output, each chunk is recorded as a new_token event containing the raw token value. These events bypass the redaction pipeline entirely — prepareRunCreateOrUpdateInputs (JS) and _hide_run_outputs (Python) only process the inputs and outputs fields on a run, never the events array. As a result, applications relying on output redaction to prevent sensitive LLM output from being stored in LangSmith will still leak the full streamed content via run events. Version 0.5.19 of the JavaScript SDK and version 0.7.31 of the Python SDK fix the issue.","ecosystems":["PyPI","npm"],"exploits":{"exploitdb":0,"first_seen":null,"github_poc":0,"max_poc_stars":null,"metasploit":0,"nuclei":0,"refs":[],"tier":"none_known"},"has_fix":true,"headline":"LangSmith SDK: Streaming token events bypass output redaction","id":"CVE-2026-41182","indexable":true,"kev":null,"kind":"cve","links":{"ecosystem":"PyPI","product":null,"products":[],"type":"information_disclosure","year":"2026"},"n_packages":5,"n_references":1,"next":{"headline":"FreeScout allows non-folder conversation queries to disclose assigned-only hidden conversations","id":"CVE-2026-41183"},"packages":[{"ecosystem":"PyPI","fixed":["0.7.31"],"introduced":["0"],"last_affected":[],"package":"langsmith","purl":null,"source":"GHSA-rr7j-v2q5-chgv"},{"ecosystem":"PyPI","fixed":["0.7.31"],"introduced":["0"],"last_affected":[],"package":"langsmith","purl":"pkg:pypi/langsmith","source":"GHSA-rr7j-v2q5-chgv"},{"ecosystem":"PyPI","fixed":["0.7.31"],"introduced":["0"],"last_affected":[],"package":"langsmith","purl":"pkg:pypi/langsmith","source":"PYSEC-2026-2583"},{"ecosystem":"npm","fixed":["0.5.19"],"introduced":["0"],"last_affected":[],"package":"langsmith","purl":null,"source":"GHSA-rr7j-v2q5-chgv"},{"ecosystem":"npm","fixed":["0.5.19"],"introduced":["0"],"last_affected":[],"package":"langsmith","purl":"pkg:npm/langsmith","source":"GHSA-rr7j-v2q5-chgv"}],"prev":{"headline":"Traefik: Errors middleware forwards Authorization and Cookie headers to separate error page service","id":"CVE-2026-41181"},"products":["langsmith-sdk"],"published":"2026-04-23","related":[],"sources":[{"class":"attribution","licence":"CVE Terms of Use (MITRE)","name":"CVE List","prefix":"CVE"},{"class":"attribution","licence":"CC BY 4.0","name":"GitHub Advisory Database","prefix":"GHSA"},{"class":"attribution","licence":"CC BY 4.0","name":"PyPA Advisory Database","prefix":"PYSEC"},{"class":"attribution","licence":"CWE Terms of Use","name":"CWE (MITRE)"},{"class":"attribution","licence":"FIRST CVSS licence","name":"CVSS (FIRST)"}],"ssvc":{"automatable":"yes","exploitation":"none","technical_impact":"partial"},"state":"PUBLISHED","updated":"2026-04-23","vendors":["langchain-ai"]},"summary":"The LangSmith SDK fails to redact sensitive data in streaming token events, allowing attackers or unauthorized users to view raw LLM outputs.","url":"https://insights.sbomify.com/CVE-2026-41182/"}