TL;DR: Cline’s safety checks can fail open when attacker-controlled repo content steers the agent into shell execution, Manifold Security documents. Approval gates need to be tied to the actual execution boundary, or developers and their reachable systems stay exposed.
Editorial analysis by NHI Mgmt Group, based on content published by Manifold Security: “CVE-2026-52024, CVE-2026-52025: We Bypassed Two of Cline's Safety Checks to Run Code (4.2M Installs)”.
Key questions
Q: Why do AI coding agents create confused-deputy risk in developer workflows?
A: Because they execute under the developer’s identity while reading attacker-controlled instructions from repositories, issues, or web content.
Q: What signs suggest an agentic approval flow is not actually controlling execution?
A: If a preview action can trigger a shell, if the model can mark its own command safe, or if the same workflow succeeds with no meaningful human gate, the control is cosmetic.
Q: What breaks when an AI coding editor can execute commands directly?
A: The control model breaks because command execution moves from human review to machine action.
Practitioner guidance
- Audit approval surfaces against the real execution sink Trace every approve, deny, preview, and inspect interaction to the underlying process it triggers, then remove any UI path that can reach a shell or interpreter unexpectedly.
- Separate model confidence from execution permission Require deterministic policy checks on the command string and its arguments before dispatch, even when the model marks the action as low risk.
- Constrain repository-driven setup flows Treat untrusted repositories as hostile inputs and limit what the agent can invoke during setup, especially when the workflow inherits developer credentials.
Bottom line: This case shows that agentic coding tools can turn ordinary repository setup into arbitrary code execution when approval gates are tied to the wrong boundary.
Glossary terms: Confused Deputy, Agentic AI, Runtime Trust Boundary, Indirect Prompt Injection
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
Approval gates that depend on the agent’s own judgment are not controls, they are acknowledgements. When the same runtime that consumes attacker-controlled text also decides whether its own command is safe, the approval model has already failed. This is not a tuning problem. It is a trust-placement problem, and practitioners should stop treating model self-attestation as a governance boundary.
A few things that frame the scale:
- Internal repositories are 6x more likely to contain secrets than public ones (32.2% vs 5.6%), contradicting the assumption that private repos are safe, according to the State of Secrets Sprawl 2026.
- A Harris Poll survey of more than 300 technology decision-makers found that 86% expect agentic AI to deliver positive ROI, yet fewer than half had AI governance policies in place.
A question worth separating out:
Q: Should organisations treat AI coding agents like privileged software identities?
A: Yes. If an AI agent can install code, access secrets, or modify repositories, it is functionally acting as a privileged non-human identity and should be governed that way. That means task-scoped access, explicit boundaries, and monitoring of its downstream actions, especially when it can touch build and release systems.
👉 Read our full editorial: Cline’s approval gap shows where agentic controls fail at runtime