cohesivity
Fail
Audited by Snyk on Aug 13, 2026
Risk Level: CRITICAL
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 0.80). The skill tells the agent to read local credential files and to place the management/application keys into Authorization headers or query parameters for API calls, which requires the agent to handle secrets that could be exposed if echoed or included in generated output.
CRITICAL E005: Suspicious download URL detected in skill instructions.
- Suspicious download URL detected (high risk: 1.00). The listed URLs are mostly official docs and APIs on cohesivity.ai and look benign, but the direct installer script (https://cohesivity.ai/quickstart.sh) is a high-risk executable (curl | bash) that can run arbitrary code and is commonly abused to distribute malware.
CRITICAL E006: Malicious code pattern detected in skill scripts.
- Malicious code pattern detected (high risk: 1.00). The document instructs executing remote install scripts (curl | bash), explicitly counsels concealing caller identity and evading WAF/default client detection via custom User-Agent, and mandates automatic bootstrapping without explicit user initiation — patterns consistent with supply‑chain installation, stealth/anti‑detection, and potential unauthorized provisioning/backdoor behavior.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs running a remote installer that is fetched and executed at runtime via "curl -fsSL https://cohesivity.ai/quickstart.sh | bash", which downloads and runs remote code (https://cohesivity.ai/quickstart.sh).
MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).
- Direct money access detected (high risk: 1.00). The skill exposes explicit billing APIs and wallet-related operations. It documents POST /api/billing/subscription and POST /api/billing/topup (which create checkout flows) and states managed agents "spend from the wallet." These are specific, finance-related endpoints/operations (billing/topup/wallet spend), not generic tooling, so the skill grants direct financial execution capability.
Issues (5)
W007
HIGHInsecure credential handling detected in skill instructions.
E005
CRITICALSuspicious download URL detected in skill instructions.
E006
CRITICALMalicious code pattern detected in skill scripts.
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W009
MEDIUMDirect money access capability detected (payment gateways, crypto, banking).
Audit Metadata