{"id":"CVE-2026-56696","aliases":["GHSA-24cw-228q-3rx9","PYSEC-2026-3880"],"title":"OpenHarness remote project-context commands allow persistent prompt poisoning","summary":"OpenHarness remote project-context commands allow persistent prompt poisoning","severity":"medium","cvss":5.4,"cvssVector":"CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:N","vendor":"openharness-ai","product":"openharness-ai","ecosystem":"pip","affected":["openharness-ai <= 0.1.9"],"published":"2026-06-23","updated":"2026-09-10","sourceUpdated":"2026-09-10T12:25:37.097640108Z","source":"OSV","sourceUrl":"https://osv.dev/vulnerability/GHSA-24cw-228q-3rx9","references":[{"url":"https://nvd.nist.gov/vuln/detail/CVE-2026-56696"},{"url":"https://github.com/HKUDS/OpenHarness/pull/272"},{"url":"https://github.com/HKUDS/OpenHarness/commit/27bb93b810e9ea8fa4832eab7152eeb3b4a6bffb"},{"url":"https://github.com/HKUDS/OpenHarness"},{"url":"https://www.vulncheck.com/advisories/openharness-prompt-injection-via-issue-and-pr-comments-slash-commands"},{"url":"https://pypi.org/project/openharness-ai"},{"url":"https://github.com/advisories/GHSA-24cw-228q-3rx9"}],"tags":["osv","pip"],"epss":0.0037,"epssPercentile":0.30916,"ingestedAt":"2026-09-04T19:32:57.192Z","slug":"CVE-2026-56696","body":"## Overview\n\nOpenHarness /issue and /pr_comments slash commands lack remote_invocable=False protection, allowing remote channel senders to write attacker-controlled Markdown into project context files. Admitted remote attackers can inject malicious content into .openharness/issue.md and .openharness/pr_comments.md files, which are subsequently injected into runtime system prompts, persistently influencing local agent behavior.\n\n## Affected packages\n\n- `openharness-ai <= 0.1.9`\n\n## Remediation\n\nRefer to the advisory for the patched release.","depth":"sunlit","depthScore":30,"depthScoreParts":{"impact":29.7,"likelihood":0.1,"exploitation":0,"ransomware":0},"changes":[]}