Manage PocketSmith transactions, categories, budgets and financial data

• Outcome: Pass
• Latest audit: May 28, 2026, 6:48 AM
• Version:
  v1.0.0

Overview

This is a disclosed PocketSmith API client for financial data, with sensitive read access and optional writes that are disabled by default.

Install only if you trust this repository with access to your PocketSmith data. Keep POCKETSMITH_ALLOW_WRITES unset unless you are actively making changes, and confirm transaction/category IDs before update or delete operations because deletions and recategorization can affect real financial records.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The README states that Claude Code will 'automatically discover and use this skill when relevant' without defining narrow trigger conditions or requiring explicit confirmation before accessing financial data. For a skill that can read sensitive transaction history and, when enabled, perform writes, broad auto-invocation language increases the chance of unintended activation and privacy-impacting actions.

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The natural-language examples are very broad ('Show me my PocketSmith transactions', 'Categorize transaction ...') and encourage the agent to infer when to invoke the skill from ordinary conversation. In a financial context, that can cause overbroad access to account data or accidental categorization changes if the agent triggers the skill without a clear, bounded user request.

Missing User Warnings

Medium

• Confidence: 86% confidence
• Finding: The skill documents delete operations for transactions and categories without an explicit warning that these actions may be irreversible or cause financial-data loss. In a finance skill, users may run examples directly, and the presence of a write-enable flag reduces accidental writes but does not mitigate the risk once writes are enabled.

VirusTotal

64/64 vendors flagged this skill as clean.