{"answers":{"affected":"2 open source packages ship an affected version, including openclaude. It affects 1 product.","exploited":"It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-42074 exists.","fix":"Upgrade openclaude to 0.5.1 or later.","what":"CVE-2026-42074 is a critical vulnerability in Gitlawb openclaude, published on 2 Jun 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-2026-42074 and fix it if it is affected.\n\nWhat it is: CVE-2026-42074 is a critical vulnerability in Gitlawb openclaude, published on 2 Jun 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.\nExploitation: It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-42074 exists.\n\nAffected packages and fixed versions:\n- openclaude (npm): affected all versions; fixed in 0.5.1\n- openclaude (npm): affected all versions; fixed in 0.5.1\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-2026-42074/ (Markdown: https://insights.sbomify.com/CVE-2026-42074/index.md)","id":"CVE-2026-42074","markdown":"https://insights.sbomify.com/CVE-2026-42074/index.md","name":"CVE-2026-42074","record":{"aliases":["GHSA-m77w-p5jj-xmhg"],"analysis":{"component_type":"developer_tool","fix_type":"upgrade","impacts":["code_execution"],"network_reachable":"yes","non_default_config":false,"product_category":"OpenClaude CLI","requires_auth":"none","summary":"A prompt-injected LLM can bypass the sandbox by setting a dangerous parameter, leading to arbitrary host-level code execution.","user_interaction":"not_required","vuln_class":"security_feature_bypass"},"analysis_generated_by":"qwen3.8-flash-next","assigner":"GitHub_M","cpe":[{"product":"openclaude","vendor":"gitlawb"}],"cvss":{"score":9.3,"severity":"critical","source":"cna","vector":"CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N","version":"4.0"},"cwes":[{"id":"CWE-306","name":"Missing Authentication for Critical Functionality"},{"id":"CWE-284","name":"Improper Access Control"}],"description":"OpenClaude is an open-source coding-agent command line interface for cloud and local model providers. Prior to version 0.5.1, the dangerouslyDisableSandbox parameter is exposed as part of the BashTool input schema, meaning the LLM (an untrusted principal per the project's own threat model) can set it to true in any tool_use response. Combined with the default allowUnsandboxedCommands: true setting, a prompt-injected model can escape the sandbox for any arbitrary command, achieving full host-level code execution. This issue has been patched in version 0.5.1.","ecosystems":["npm"],"exploits":{"exploitdb":0,"first_seen":null,"github_poc":0,"max_poc_stars":null,"metasploit":0,"nuclei":0,"refs":[],"tier":"weaponised"},"has_fix":true,"headline":"OpenClaude: Sandbox Bypass via Model-Controlled `dangerouslyDisableSandbox` Input","id":"CVE-2026-42074","indexable":true,"kev":null,"kind":"cve","links":{"ecosystem":"npm","product":null,"products":[],"type":"security_feature_bypass","year":"2026"},"n_packages":2,"n_references":3,"next":{"headline":"Evolver: Path Traversal via `--out` flag in `fetch` command allows Arbitrary File Write","id":"CVE-2026-42075"},"packages":[{"ecosystem":"npm","fixed":["0.5.1"],"introduced":["0"],"last_affected":[],"package":"openclaude","purl":null,"source":"GHSA-m77w-p5jj-xmhg"},{"ecosystem":"npm","fixed":["0.5.1"],"introduced":["0"],"last_affected":[],"package":"openclaude","purl":"pkg:npm/openclaude","source":"GHSA-m77w-p5jj-xmhg"}],"prev":{"headline":"OpenClaude's MCP OAuth Callback: State Check Bypass via error Param Leads to DoS","id":"CVE-2026-42073"},"products":["openclaude"],"published":"2026-06-02","related":[{"cvss_score":8.4,"headline":"OpenClaude has Sandbox Bypass via Early-Exit Logic Flaw that Allows Path Traversal","id":"CVE-2026-35570","kev_date_added":null,"published":"2026-04-20"},{"cvss_score":6.5,"headline":"OpenClaude's MCP OAuth Callback: State Check Bypass via error Param Leads to DoS","id":"CVE-2026-42073","kev_date_added":null,"published":"2026-06-02"}],"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":"US public information, NVD API terms","name":"NVD"},{"class":"attribution","licence":"CWE Terms of Use","name":"CWE (MITRE)"},{"class":"attribution","licence":"FIRST CVSS licence","name":"CVSS (FIRST)"}],"ssvc":{"automatable":"yes","exploitation":"poc","technical_impact":"total"},"state":"PUBLISHED","updated":"2026-06-02","vendors":["Gitlawb"]},"summary":"A prompt-injected LLM can bypass the sandbox by setting a dangerous parameter, leading to arbitrary host-level code execution.","url":"https://insights.sbomify.com/CVE-2026-42074/"}