Persistent prompt injection via remote context commands
Published Jun 23, 2026 · Updated Jun 23, 2026
Prompt injection in OpenHarness 0.1.9 and earlier allows remote authenticated users to persist attacker-controlled agent context. The /issue and /pr_comments handlers inherit remote_invocable=True and write supplied Markdown to .openharness/issue.md and .openharness/pr_comments.md, which runtime prompt assembly later loads. Access requires admission by the configured remote channel policy; injected context can influence later local agent behavior but does not directly execute commands.
Summary
What happened
Prompt injection in OpenHarness 0.1.9 and earlier allows remote authenticated users to persist attacker-controlled agent context. The /issue and /pr_comments handlers inherit remote_invocable=True and write supplied Markdown to .openharness/issue.md and .openharness/pr_comments.md, which runtime prompt assembly later loads. Access requires admission by the configured remote channel policy; injected context can influence later local agent behavior but does not directly execute commands.
The record
- CVE
- CVE-2026-56696
- Published
- Jun 23, 2026
- Updated
- Jun 23, 2026
- Vendor
- Data Intelligence Lab @ HKU
- Product
- OpenHarness
- Classifications
- CWE-862, T1565.001
- Attack vector
- network
- Privileges
- authenticated
Timeline
How it unfolded
- Jun 23, 2026CVE publishedPublication date reported by the CVE source.
- Jun 23, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=0 <=0.1.9
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Public exploit references
- Upstream remote project-context poisoning reproductionproof of concept · demonstrated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo