AEV-2026-0003
Strands Agents Tools shell consent gate skippable by the model (CVE-2026-18733)
AWS Strands Agents Tools shipped a shell tool whose human consent gate was skipped when a model settable non_interactive parameter was true, letting crafted input run arbitrary operating system commands with the agent process's privileges. Disclosed by AWS, patched in 0.8.0, with no evidence of exploitation.
Affected
- Organisation
- AWS
- Product
- Strands Agents Tools
- Component
- shell tool consent gate (non_interactive parameter)
- Versions
- Versions before 0.8.0; fixed in 0.8.0
- Configurations
- Agent exposed to untrusted content that can influence tool parameters
Execution authority facts
Fields the evidence does not establish are shown as unknown rather than guessed.
- Intended task
- An agent running operating system commands through the shell tool
- Agent
- An agent built on the Strands Agents SDK
- Delegating actor
- The operator running the agent
- Action
- Set non_interactive to true so the consent gate did not fire, then ran an operating system command
- Target resource
- The host the agent process runs on
- Environment
- Wherever the agent process executes
- Credentials used
- unknown
- Privileges available
- The privileges of the agent process
- Authority presented
- A tool call that also carried the parameter governing its own approval
- Authority required
- Operator consent to run the command
- Applicable policy
- The human consent gate on the shell tool
- Approval mechanism
- A consent gate the operator was meant to satisfy
- Required approver
- The operator
- Independent approval
- no
- Action binding
- The gate's enabling condition was a model controllable field, not operator set
- Sequence context
- Indirect prompt injection in content the agent read could set the parameter before the command ran
Impact
- Consequence
- Arbitrary operating system commands run with the agent process's privileges
- Reach
- The host the agent runs on
- Reversibility
- unknown
- Detectability
- Silent: the consent gate did not fire
- Propagation
- unknown
- Recovery
- Fixed by removing the parameter from the model facing schema and reading it from an environment variable
Evidence
Primary sources
- Prompt injection bypasses shell tool consent gate in Strands Agents Tools (AWS, strands-agents/tools (GitHub Security Advisory GHSA-mqvc-p852-wf8x))
Supporting sources
- CVE-2026-18733 - AWS Security Bulletin 2026-072-AWS (AWS)
- strands-agents/tools, src/strands_tools/shell.py (current main branch) (GitHub, strands-agents/tools)
- Reproduction status
- Confirmed and patched by AWS; CVSS 3.1 base 8.8 in the GitHub advisory
- Evidence state
- Confirmed
Known unknowns
- Whether any deployment was affected before the patch.
Limitations
- No evidence in AWS's advisories, GitHub's advisories or public reporting that it was exploited against a real deployment.
- Three sibling CVEs in the same package address a related shape in credential scope, tenant isolation and connection configuration and are recorded on the source record rather than folded in here.
Claim provenance
- verified
AWS assigned CVE-2026-18733; the GitHub advisory records the exploit shape, affected and fixed versions and a CVSS 3.1 base of 8.8.
The Agent Had to Ask Permission. It Could Also Turn Permission Off.
