{"answers":{"affected":"2 open source packages ship an affected version, including openclaw. It affects 1 product.","exploited":"There is no known exploitation of CVE-2026-27001 in the wild and no public exploit code.","fix":"Upgrade openclaw to 2026.2.15 or later.","what":"CVE-2026-27001 is a high-severity code injection vulnerability in Openclaw Openclaw, published on 19 Feb 2026. It can only be exploited from the same machine or local network, with no action from the victim."},"fix_prompt":"Check this codebase for CVE-2026-27001 and fix it if it is affected.\n\nWhat it is: CVE-2026-27001 is a high-severity code injection vulnerability in Openclaw Openclaw, published on 19 Feb 2026. It can only be exploited from the same machine or local network, with no action from the victim.\nExploitation: There is no known exploitation of CVE-2026-27001 in the wild and no public exploit code.\n\nAffected packages and fixed versions:\n- openclaw (npm): affected all versions; fixed in 2026.2.15\n- openclaw (npm): affected all versions; fixed in 2026.2.15\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-2026-27001/ (Markdown: https://insights.sbomify.com/CVE-2026-27001/index.md)","id":"CVE-2026-27001","markdown":"https://insights.sbomify.com/CVE-2026-27001/index.md","name":"CVE-2026-27001","record":{"aliases":["GHSA-2qj5-gwg2-xwc4"],"analysis":{"component_type":"desktop_application","fix_type":"upgrade","impacts":["code_execution","security_bypass"],"network_reachable":"adjacent_or_local","non_default_config":false,"product_category":"AI assistant","requires_auth":"none","summary":"Attackers can inject malicious instructions into LLM prompts via unsanitized directory names, potentially executing arbitrary code or bypassing security controls.","user_interaction":"not_required","vuln_class":"code_injection"},"analysis_generated_by":"qwen3.8-flash-next","assigner":"GitHub_M","cpe":[{"product":"openclaw","vendor":"openclaw"}],"cvss":{"score":8.6,"severity":"high","source":"cna","vector":"CVSS:4.0/AV:L/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-77","name":"Improper Neutralization of Special Elements used in a Command"}],"description":"OpenClaw is a personal AI assistant. Prior to version 2026.2.15, OpenClaw embedded the current working directory (workspace path) into the agent system prompt without sanitization. If an attacker can cause OpenClaw to run inside a directory whose name contains control/format characters (for example newlines or Unicode bidi/zero-width markers), those characters could break the prompt structure and inject attacker-controlled instructions. Starting in version 2026.2.15, the workspace path is sanitized before it is embedded into any LLM prompt output, stripping Unicode control/format characters and explicit line/paragraph separators. Workspace path resolution also applies the same sanitization as defense-in-depth.","ecosystems":["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":"OpenClaw: Unsanitized CWD path injection into LLM prompts","id":"CVE-2026-27001","indexable":true,"kev":null,"kind":"cve","links":{"ecosystem":"npm","product":"openclaw:openclaw","products":["openclaw:openclaw"],"type":"code_injection","year":"2026"},"n_packages":2,"n_references":3,"next":{"headline":"OpenClaw: Docker container escape via unvalidated bind mount config injection","id":"CVE-2026-27002"},"packages":[{"ecosystem":"npm","fixed":["2026.2.15"],"introduced":["0"],"last_affected":[],"package":"openclaw","purl":null,"source":"GHSA-2qj5-gwg2-xwc4"},{"ecosystem":"npm","fixed":["2026.2.15"],"introduced":["0"],"last_affected":[],"package":"openclaw","purl":"pkg:npm/openclaw","source":"GHSA-2qj5-gwg2-xwc4"}],"prev":{"headline":"Traefik: tcp router clears read deadlines before tls forwarding, enabling stalled handshakes (slowloris doS)","id":"CVE-2026-26999"},"products":["openclaw"],"published":"2026-02-19","related":[{"cvss_score":9.9,"headline":"Security feature bypass in OpenClaw","id":"CVE-2026-28363","kev_date_added":null,"published":"2026-02-27"},{"cvss_score":9.8,"headline":"Code injection in n/a","id":"CVE-2026-30741","kev_date_added":null,"published":"2026-03-11"},{"cvss_score":9.4,"headline":"OpenClaw \u003c 2026.3.28 - Privilege Escalation via Missing Caller Scope Validation in Device Pair Approval","id":"CVE-2026-33579","kev_date_added":null,"published":"2026-03-31"},{"cvss_score":9.4,"headline":"OpenClaw \u003c 2026.3.11 - Privilege Escalation via Unvalidated Scope in device.token.rotate","id":"CVE-2026-32922","kev_date_added":null,"published":"2026-03-29"},{"cvss_score":9.4,"headline":"OpenClaw \u003c 2026.3.11 - Approval Bypass via Unrecognized Script Runners","id":"CVE-2026-32978","kev_date_added":null,"published":"2026-03-29"},{"cvss_score":9.4,"headline":"OpenClaw \u003c 2026.3.12 - Scope Elevation in WebSocket Shared-Auth Connections","id":"CVE-2026-22172","kev_date_added":null,"published":"2026-03-20"},{"cvss_score":9.4,"headline":"OpenClaw \u003c 2026.2.14 - Remote Code Execution via Node Invoke Approval Bypass","id":"CVE-2026-28466","kev_date_added":null,"published":"2026-03-05"}],"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":"no","exploitation":"none","technical_impact":"total"},"state":"PUBLISHED","updated":"2026-02-20","vendors":["openclaw"]},"summary":"Attackers can inject malicious instructions into LLM prompts via unsanitized directory names, potentially executing arbitrary code or bypassing security controls.","url":"https://insights.sbomify.com/CVE-2026-27001/"}