Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessIdentify network-accessible miniclawd instance
-
DeliverySend crafted request with shell metacharacters in requires.bins
-
ExploitSkillsLoader which() passes unsanitized input to shell
-
ExecutionInjected commands execute in miniclawd process context
-
ImpactPartial host access achieved (data read/write, process disruption)
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A05:2025 - Injection
ASVS 5.0
V1 - Encoding and Sanitization Command injection prevention
MITRE ATT&CK
1 technique across 1 tactic: Execution The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | CVSS 4.0 vector AV:N/AC:L/AT:N/PR:N/UI:N confirms: No special conditions - remote unauthenticated exploitation against default configurations of miniclawd. The SkillsLoader component must be active and processing attacker-reachable input (i.e., miniclawd must be network-accessible and the skill-loading pathway must be reachable without authentication). Limiting factors are: (1) miniclawd is a relatively niche project, limiting the attack surface to organizations that have specifically deployed it; (2) technical impact is partial (VC:L/VI:L/VA:L) suggesting process-level rather than full system compromise, though this depends on the privilege level of the miniclawd process; (3) no active mass exploitation confirmed (no KEV listing). The rolling release model means no configuration-specific mitigations are documented by the vendor. |
| Impact | Despite a moderate CVSS 4.0 base score of 5.5, the attack profile warrants elevated attention due to converging risk signals. The vector AV:N/AC:L/AT:N/PR:N/UI:N represents a worst-case accessibility profile - remotely exploitable, low complexity, no authentication, no user interaction - which is inconsistent with a 'moderate' risk framing and primarily reflects the partial confidentiality/integrity/availability impact (VC:L/VI:L/VA:L) rather than any meaningful barrier to exploitation. EPSS at 1.04% (78th percentile) indicates elevated exploitation probability relative to the CVE corpus, though mass scanning campaigns are not yet confirmed. SSVC flags this as automatable with a POC exploitation state, reinforcing that scripted attacks are feasible. The absence of KEV listing is the primary mitigating signal - no confirmed active exploitation by CISA at time of analysis. A critical compounding factor is the vendor's non-response to disclosure and the rolling-release model, which means defenders cannot rely on a patched release version. The combination of a public POC, automatable attack, zero authentication barrier, and no patch creates a materially higher real-world risk than the CVSS score alone implies. |
| Exploit ScenarioDerived | A remote unauthenticated attacker identifies a network-accessible miniclawd instance and submits a crafted request that populates the |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Jun 08, 2026
Command injection in FoundDream miniclawd's SkillsLoader component exposes systems to remote unauthenticated arbitrary command execution via unsanitized manipulation of the requires.bins argument in /src/application/skills-loader.ts. The CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:N/UI:N) confirms zero-barrier remote exploitation with no authentication or user interaction required, and SSVC classifies the attack as automatable with partial technical impact. A public exploit (POC) exists via GitHub issue tracker, the vendor has not responded to disclosure, and no patch has been released - leaving affected deployments with no official remediation path.
RemediationAI
No vendor-released patch identified at time of analysis. The FoundDream project was notified via issue report but has not responded, and no remediated commit or release is available. Organizations running miniclawd should treat this as an unpatched vulnerability with a public exploit. Specific compensating controls in priority order: First, restrict network access to miniclawd instances using firewall rules or network segmentation - limit exposure to trusted internal hosts only, accepting the trade-off of reduced accessibility. Second, if the SkillsLoader component or binary resolution feature is not required for your use case, disable or remove the /src/application/skills-loader.ts module from deployment, noting this may break skill-loading functionality. Third, implement input validation at the application boundary to strip or reject shell metacharacters (;, |, ` `, $(), &&, ||) from any requires.bins values before they reach the which` function - this requires a code-level change and carries risk of breaking legitimate binary names if over-aggressively filtered. Monitor the GitHub repository at https://github.com/FoundDream/miniclawd/ and issue tracker for any vendor response or upstream patch.
Affected VersionsENISA
Affects your stack
Find out whether the stacks and servers you run are on an affected version.
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
miniclawd is a TypeScript-based application by FoundDream that implements a SkillsLoader component responsible for loading skills and resolving binary dependencies. The vulnerable which function in /src/application/skills-loader.ts accepts a requires.bins argument - likely a list of binary names to locate on the host system - and passes this value unsanitized into a shell command execution context. CWE-77 (Improper Neutralization of Special Elements used in a Command) is the root cause class: attacker-controlled input containing shell metacharacters (e.g., semicolons, backticks, pipes) is interpreted by the underlying shell rather than treated as a literal string. The CPE cpe:2.3:a:founddream:miniclawd:*:*:*:*:*:*:*:* with a wildcard version indicates all tracked releases are affected up through commit 2d65665046e2222eeea76cafc8570ed546a8c125.
Know when Miniclawd gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P
Primary rating from Vendor (VulDB) · only source for this CVE.
CVSS VectorVendor: VulDB
Lifecycle Timeline
2DescriptionCVE.org
A vulnerability was detected in FoundDream miniclawd up to 2d65665046e2222eeea76cafc8570ed546a8c125. This affects the function which of the file /src/application/skills-loader.ts of the component SkillsLoader. Performing a manipulation of the argument requires.bins results in command injection. The attack may be initiated remotely. The exploit is now public and may be used. This product uses a rolling release model to deliver continuous updates. As a result, specific version information for affected or updated releases is not available. The project was informed of the problem early through an issue report but has not responded yet.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
OS command injection occurs when an application passes unsanitized user input directly into system shell commands.
EUVD-2026-31671
GHSA-cjqg-8xv4-hwwm
NVD
CVE.org
EPSS
Vulnrichment
VulDB
Exploit-DB