CVE-2026-45374: DeepSeek TUI: task_create Insecure Defaults Enable RCE via Prompt Injection in Project Files

Should you act?

Critical and remotely reachable

No known exploitation yet, but it is network-reachable without credentials. This signal is computed from CISA KEV, EPSS and CVSS data; it is not an editorial verdict.

• Not in CISA KEV, but CISA’s SSVC assessment records a public proof-of-concept, so attack code exists.
• EPSS 0.5%, percentile 39: FIRST’s model estimates a 0.5% chance of exploitation activity in the next 30 days, higher than 39% of all scored CVEs (low). EPSS is a probability, not a verdict.
• Needs a user to click or open something. Phishing makes that realistic, but it is not wormable.
• Only applies if your dependency tree includes deepseek-tui (crates.io (Rust)) < 0.8.26, fixed in 0.8.26. Check your lockfile, including transitive dependencies.

What it is

CodeWhale is a DeepSeek + MiMo coding agent in terminal. Prior to 0.8.26, the taskcreate tool spawns durable sub-agents that inherit two insecure defaults, allowshell defaults to true (config.rs:1499: self.allowshell.unwrapor(true)) and autoapprove defaults to true (taskmanager.rs:297: autoapprove: Some(true)). When a user approves a taskcreate call (which requires ApprovalRequirement::Required), they approve what appears to be a benign work prompt. However, the spawned sub-agent silently receives unrestricted, unapproved shell access. This vulnerability is fixed in 0.8.26.

From the GitHub advisory

Summary

The taskcreate tool spawns durable sub-agents that inherit two insecure defaults:

- allowshell defaults to true (config.rs:1499: self.allowshell.unwrapor(true)) - autoapprove defaults to true (taskmanager.rs:297: autoapprove: Some(true))

When a user approves a taskcreate call (which requires ApprovalRequirement::Required), they approve what appears to be a benign work prompt (e.g., "fix TODOs and write a README"). However, the spawned sub-agent silently receives unrestricted, unapproved shell access. Neither allowshell nor autoapprove need to be explicitly specified by the model and both default to true.

PoC Step 1 — Create a malicious repo:

Weakness type: CWE-94.

Product: deepseek-tui (crates.io (Rust)) · Assigned by: GitHub_M · Published: 2026-05-28

Affected and fixed versions

Severity scores

• Attack vector: network (remotely reachable)
• Attack complexity: low
• Privileges required: none (no login)
• User interaction: required (a user must click/open something)
• Scope: changed (can affect other components)

CVSS measures how bad a flaw is if exploited, not how likely exploitation is. Read it next to the exploitation data below.

Exploitation and urgency

• CISA KEV: not listed (no confirmed in-the-wild exploitation known to CISA).
• CISA SSVC assessment: exploitation poc, automatable no, technical impact total.
• FIRST EPSS (model of 2026-10-21): 0.47% probability of exploitation activity in the next 30 days; percentile 39.

Timeline

• 2026-05-14 — GitHub-reviewed advisory GHSA-72w5-pf8h-xfp4 published (critical)
• 2026-05-28 — CVE record published by GitHub_M
• 2026-05-30 — CVE record last updated

Sources and references

Original sources first. Version and fix details should always be confirmed against the vendor or project advisory.

Vulnerability records

1. GitHub Advisory GHSA-72w5-pf8h-xfp4 github.com
2. CVE.org record CVE-2026-45374 cve.org
3. NVD entry CVE-2026-45374 nvd.nist.gov

Other references from the CVE record and advisory

1. github.com github.com
2. github.com github.com
3. github.com github.com

Critical crates.io (Rust) advisories · Today’s Security Digest