Security audit

sente

Security checks for vulnerabilities and agentic risk

Overview

This skill is purpose-aligned but asks the agent to handle powerful account credentials and sessions with some under-protected setup patterns users should review carefully.

Install only if you trust Sente and are comfortable giving it durable access to email identities and connected accounts. Prefer a pinned CLI version, a low-privilege isolated environment, protected secret storage, account-specific authorization, and immediate deletion or secure handling of exported browser state.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
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

Findings (2)

T08 · Insecure Dependencies

Error

Location: SKILL.md:19
Finding: Unpinned Global Installation of a Security-Sensitive Third-Party CLI
Content: View full analysis
Remediation: View remediation
``` 2. Publish and verify package integrity, provenance, and signatures through an appropriate trusted mechanism. 3. Document the official npm organization and package ownership so users can detect typosquatting or ownership changes. 4. Avoid global installation where possible. Prefer an isolated project environment, container, or other sandbox with minimum filesystem and network privileges. 5. Disable unnecessary npm lifecycle scripts during installation where compatible with the package. 6. Review the CLI and its complete dependency tree before updating the pinned version. 7. Establish a controlled update process that performs security review, tests credential-handling behavior, and records approved hashes before deployment. ]]>

T09 · Insecure Skill Coding Practices

Error

Location: SKILL.md:37
Finding: Secrets Exposed Through Command-Line Arguments and Unprotected Files
Content: View full analysis
> echo "SENTE_IDENTITY_ID=" >> ``` ``` ```markdown ```bash sente connect https://app.example.com/login --identity \ --username --password \ --totp-seed # optional: base32 or otpauth:// URI sente connections # list connected accounts sente connection revoke # stop using it (vault kept; reconnect re-enables) sente connection delete # revoke AND purge the stored credentials sente session export ./state.json # Playwright storageState for your own browser stack ``` ``` ### Technical Analysis The documented `sente connect` invocation passes passwords and TOTP seeds directly through command-line arguments. Depending on the operating system, shell, process-monitoring configuration, and agent runtime, these values may be exposed through: - Process listings or process inspection interfaces. - Shell command history. - Terminal or orchestration logs. - Agent transcripts and debugging telemetry. - Audit systems that record command invocations. A TOTP seed is especially sensitive because it permits generation of future authentication codes, rather than exposing only one short-lived code. The setup also appends `SENTE_API_TOKEN` to an unspecified environment file without requiring restrictive file permissions, a protected location, or exclusion from version control and backups. The command substitutes the token into a shell command, which may additional ...[truncated 2882 chars]
Remediation: View remediation

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.