CVE
CodeWhale auth bypass in automation auto-approval
Affects CodeWhale 0.8.67 and earlier, where an omitted auto_approve on an automation resolves to true. Fixed in 0.9.0.
Written by anthropic/claude-haiku-5.5 from the sources below. Verify against them.
CVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:U/C:H/I:H/A:H
- Exploited
- No known exploitation
- Fix
- Fixed in 0.9.0
- Who is exposed
- Specific setups only
Do first: Check whether your setup matches what the bug needs. CodeWhale 0.8.67 and earlier are affected (npm codewhale and codewhale-tui crate). An omitted auto_approve on an automation resolves to true. The automation_create and automation_run tools are registered by default.
- Last modified
- Whether it needs action now or is a reference record, by fixed rules, with why. Last checked is when cve.now last re-read it.What do these mean? →
- Reference
- Last checked
- A 0-100 score for how much a CVE matters across the industry: exploitation, exposure, how widely the software is used, severity and whether a fix exists.What do these mean? →
- 21/100 · provisional
- How many real-world setups can hit the bug, from the vendor's advisory and CISA data. It does not know your setup.What do these mean? →
- Specific setups only
- Whether working attack code is public (Exploit-DB, Metasploit, Nuclei or GitHub). Public code makes attacks easier.What do these mean? →
- None on Exploit-DB, Metasploit or GitHub
- Exploitation: none
- Automatable: no
- Technical impact: total
Summary
CodeWhale, a terminal coding agent, can be tricked into silently approving later scheduled runs that use mutating tools, because automation auto-approval defaults to true. A remote attacker without a login could exploit this after a user approves an apparently benign recurring automation, and an automation with shell access could then repeatedly run commands.
Original description
CodeWhale is a DeepSeek and MiMo coding agent for the terminal. Prior to 0.9.0, DEFAULT_AUTOMATION_AUTO_APPROVE caused AutomationRecord::task_auto_approve to resolve an omitted setting to true, and the automation_create schema advertised the same insecure default. enqueue_run_task converted that omission into an explicit true value, bypassing the safer task-path fallback and selecting ApprovalMode::Bypass for each scheduled run. After a user approved creation of an apparently benign recurring automation, subsequent executions could silently authorize mutating tools without a human present, and an automation created with shell access could repeatedly execute commands. This issue is fixed in version 0.9.0.
Who should check
Teams running CodeWhale 0.8.67 or earlier with the automation_create tool, the DeepSeek and MiMo terminal coding agent.
“CodeWhale 0.8.67 and earlier are affected.”
cve.now Index
- Exploitation0/40No signal
- Exposure12/25Specific setups only, reachable over the network without a login
- Reach?/20Unknown. No tracked technology or package matched
- Severity8.5/10CVSS 7.5, total technical impact per CISA
- No patch0/5No signal
What would change the score
- Would score about 71 if added to CISA KEV
- Would score about 66 if CISA's SSVC rates exploitation active
What to do
Check whether your setup matches what the bug needs.
CodeWhale 0.8.67 and earlier are affected (npm codewhale and codewhale-tui crate). An omitted auto_approve on an automation resolves to true. The automation_create and automation_run tools are registered by default.Upgrade to a fixed version: 0.9.0.
Check your systems with an agent
Paste this into a coding or ops agent. It checks your systems read-only and reports a table per asset.
Check whether my systems are affected by CVE-2026-83817 (CodeWhale: Automation auto_approve defaults to true, bypassing the task-path hardening fix for silent recurring agent execution), using the cve.now data below.
Rules: work read-only. Do not run exploit code, proof-of-concept scripts or scanners that send attack payloads, and do not change any configuration, package, image or deployment. Report findings only.
What cve.now has for CVE-2026-83817 (entry last updated 2026-11-16):
Affected:
- codewhale-hq CodeWhale: fixed in 0.9.0
Fixed versions: 0.9.0.
Conditions (all must be true for the bug to be hit):
- CodeWhale 0.8.67 and earlier are affected (npm codewhale and codewhale-tui crate). (source: https://github.com/codewhale-hq/Codewhale/security/advisories/GHSA-8hp3-xxrg-v9c8)
- An omitted auto_approve on an automation resolves to true. (source: https://github.com/codewhale-hq/Codewhale/security/advisories/GHSA-8hp3-xxrg-v9c8)
- The automation_create and automation_run tools are registered by default. (source: https://github.com/codewhale-hq/Codewhale/security/advisories/GHSA-8hp3-xxrg-v9c8)
Setup: the bug is reachable in a default setup; no special feature or setting is needed.
Attacker needs: network access, no login, a user action.
CISA KEV: CVE-2026-83817 is not listed as of cve.now's last read.
This KEV status is for CVE-2026-83817 only. If you also check another CVE, look up its own KEV status; never assume it from this one.
How to check:
- Installed software: check what is actually running, not what a repo declares. Read the version from each host, container image and Kubernetes node (the binary's version output, the OS package database, image layers), and check whether the conditions above hold there. Distribution packages may carry a backported fix under an older version number; confirm against the distribution's own advisory.
Verify against the linked advisories before concluding, and cite the date you read each one:
- https://github.com/codewhale-hq/Codewhale/security/advisories/GHSA-8hp3-xxrg-v9c8
- https://github.com/codewhale-hq/Codewhale/commit/208421b9d23d81cea3311a1eeb2a2ea81d637686
- https://github.com/codewhale-hq/Codewhale/releases/tag/v0.9.0
Do not conclude an asset is unaffected unless you checked it and the evidence covers both its version and the conditions above. If you could not check something, mark it unknown and say what is missing.
Report a table with one row per asset and these columns: Asset | Version or build | Status (affected, fixed, not applicable, unknown) | Compensating controls | Exposure (internet-facing, internal, none) | Evidence | Next action.
Entry: https://cve.now/CVE-2026-83817
For follow-up questions, use the cve.now MCP server at https://mcp.cve.now/mcp (no authentication), for example execute with name cve.get and arguments {"id": "CVE-2026-83817"}.Timeline
- CVE published
- First seen by cve.now
- CISA SSVC assessment
Exploitation: none · Automatable: no · Technical impact: total
- Record last updated
Affected versions
- codewhale-hq CodeWhale
fixed in 0.9.0
Evidence 5
Vendor advisory · 1
Establishes: Affected and fixed versions.
- Advisory
CVE record · 2
Establishes: The identifier and description.
- Record
- Record
Other links · 2
- Repository
- Repository
Corrections
None so far.
Something wrong on this page? Report an error. Every fix is listed here.
Questions about CVE-2026-83817
Is CVE-2026-83817 being exploited?
It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. No public exploit code is tracked.Source: cve.now feed data
What does this mean?
An automation created without an explicit auto_approve setting defaults to auto-approving tool calls. Its scheduled runs then silently approve mutating tools with no human present, and if allow_shell is set, they also run shell commands silently.Source: github.com
Which versions are affected by CVE-2026-83817?
CodeWhale: before 0.9.0.Source: cve.now feed data
Am I affected, and how do I check?
CodeWhale 0.8.67 and earlier are affected, including the npm codewhale package and the codewhale-tui crate. The automation_create and automation_run tools are registered by default, so check the installed package version and whether automations were created without an explicit auto_approve value.Source: github.com
How do I fix CVE-2026-83817?
Upgrade CodeWhale to 0.9.0 or later, which is the fixed version listed for this vulnerability.Source: github.com