{"answers":{"affected":"It affects 1 product.","exploited":"There is no known exploitation of CVE-2025-54135 in the wild and no public exploit code.","fix":"No fix has been published for CVE-2025-54135 yet. Limit exposure of the affected software until one is available.","what":"CVE-2025-54135 is a high-severity code injection vulnerability in cursor, published on 5 Aug 2025. Any signed-in user can exploit it over the network, with no action from the victim."},"fix_prompt":"Check this codebase for CVE-2025-54135 and fix it if it is affected.\n\nWhat it is: CVE-2025-54135 is a high-severity code injection vulnerability in cursor, published on 5 Aug 2025. Any signed-in user can exploit it over the network, with no action from the victim.\nExploitation: There is no known exploitation of CVE-2025-54135 in the wild and no public exploit code.\n\nAffected product: cursor. No package-level version data is available, so check the vendor advisory linked from the reference page.\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-2025-54135/ (Markdown: https://insights.sbomify.com/CVE-2025-54135/index.md)","id":"CVE-2025-54135","markdown":"https://insights.sbomify.com/CVE-2025-54135/index.md","name":"CVE-2025-54135","record":{"aliases":[],"analysis":{"component_type":"desktop_application","fix_type":"upgrade","impacts":["code_execution"],"network_reachable":"yes","non_default_config":false,"product_category":"AI code editor","requires_auth":"low_privilege","summary":"Attackers can achieve remote code execution by exploiting prompt injection to write malicious MCP configuration files without user approval.","user_interaction":"not_required","vuln_class":"code_injection"},"analysis_generated_by":"qwen3.8-flash-next","assigner":"GitHub_M","cpe":[{"product":"cursor","vendor":"anysphere"}],"cvss":{"score":8.6,"severity":"high","source":"cna","vector":"CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:C/C:H/I:H/A:H","version":"3.1"},"cwes":[{"id":"CWE-829","name":"Inclusion of Functionality from Untrusted Control Sphere"},{"id":"CWE-78","name":"Improper Neutralization of Special Elements used in an OS Command"}],"description":"Cursor is a code editor built for programming with AI. Cursor allows writing in-workspace files with no user approval in versions below 1.3.9, If the file is a dotfile, editing it requires approval but creating a new one doesn't. Hence, if sensitive MCP files, such as the .cursor/mcp.json file don't already exist in the workspace, an attacker can chain a indirect prompt injection vulnerability to hijack the context to write to the settings file and trigger RCE on the victim without user approval. This is fixed in version 1.3.9.","ecosystems":[],"exploits":{"exploitdb":0,"first_seen":"2025-08-04","github_poc":2,"max_poc_stars":0,"metasploit":0,"nuclei":0,"refs":[{"date":"2025-08-04T09:56:12Z","kind":"github_poc","ref":"https://github.com/anntsmart/CVE-2025-54135","stars":0,"title":null},{"date":"2025-08-30T00:41:29Z","kind":"github_poc","ref":"https://github.com/hn1e13/test-mcp","stars":0,"title":"A repo for CVE-2025-54135 reproduction"}],"tier":"poc_unverified"},"has_fix":null,"headline":"Cursor Agent is vulnerable to prompt injection via MCP Special Files","id":"CVE-2025-54135","indexable":true,"kev":null,"kind":"cve","links":{"ecosystem":null,"product":null,"products":[],"type":"code_injection","year":"2025"},"n_packages":0,"n_references":1,"next":{"headline":"Cursor's Modification of MCP Server Definitions Bypasses Manual Re-approvals","id":"CVE-2025-54136"},"packages":[],"prev":{"headline":"HAX CMS NodeJs's Improper Error Handling Leads to Denial of Service","id":"CVE-2025-54134"},"products":["cursor"],"published":"2025-08-05","related":[{"cvss_score":9.3,"headline":"Cursor Desktop sandbox escape via agent-controlled working directory","id":"CVE-2026-50548","kev_date_added":null,"published":"2026-06-25"},{"cvss_score":9.3,"headline":"Cursor Desktop sandbox escape via symlink and failed path canonicalization","id":"CVE-2026-50549","kev_date_added":null,"published":"2026-06-25"},{"cvss_score":8.8,"headline":"Cursor: Speedbump Modal Bypass in MCP Server Deep-Link","id":"CVE-2025-64106","kev_date_added":null,"published":"2025-11-04"},{"cvss_score":8.8,"headline":"Cursor is Vulnerable to Path Manipulation Using Backslashes on Windows","id":"CVE-2025-64107","kev_date_added":null,"published":"2025-11-04"},{"cvss_score":8.8,"headline":"Cursor's Sensitive File Modification can Lead to NTFS Path Quirks","id":"CVE-2025-64108","kev_date_added":null,"published":"2025-11-04"},{"cvss_score":8.8,"headline":"Cursor CLI's Cursor Agent MCP OAuth2 Communication is Vulnerable to Remote Code Execution","id":"CVE-2025-61591","kev_date_added":null,"published":"2025-10-03"},{"cvss_score":8.8,"headline":"Cursor CLI: Arbitrary Code Execution Possible through Permissive CLI Config","id":"CVE-2025-61592","kev_date_added":null,"published":"2025-10-03"}],"sources":[{"class":"attribution","licence":"CVE Terms of Use (MITRE)","name":"CVE List","prefix":"CVE"},{"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)"},{"class":"open","licence":"CC0","name":"PoC-in-GitHub"}],"ssvc":{"automatable":"no","exploitation":"none","technical_impact":"total"},"state":"PUBLISHED","updated":"2025-08-05","vendors":["cursor"]},"summary":"Attackers can achieve remote code execution by exploiting prompt injection to write malicious MCP configuration files without user approval.","url":"https://insights.sbomify.com/CVE-2025-54135/"}