Security audit

Pilot Model Share

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward model-sharing helper, but its example checksum verification should not be treated as strong model authenticity validation.

Install only if you already trust the Pilot Protocol setup and the peers you will exchange models with. Do not rely on the included MD5 checksum snippet as proof that a model came from a trusted source; use authenticated metadata or signatures and prefer safer model formats when receiving files from others.

A.I.G

Vulnerability Patterns

Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally

Findings (1)

T09 · Insecure Skill Coding Practices

Warning

Location: SKILL.md:51
Finding: Untrusted and Cryptographically Weak Model Checksum Verification
Content: View full analysis
Remediation: View remediation
&2 exit 1 fi if [ "$EXPECTED_CHECKSUM" != "$ACTUAL_CHECKSUM" ]; then echo "Model verification failed" >&2 exit 1 fi echo "Model verified" ``` 6. Treat model formats capable of carrying executable or serialized code as untrusted even after integrity verification. Load them in a sandbox with minimal filesystem, network, and credential access, and prefer non-executable formats such as SafeTensors where applicable. ]]>

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Static analysis

No suspicious patterns detected.