ClawHub
Back to skill

Security audit

fulcra-ingest

Security checks for vulnerabilities and agentic risk

Overview

This skill is a Fulcra data-ingestion helper with disclosed remote file, record, archive, and metadata workflows that fit its stated purpose.

Before installing, be comfortable letting the agent use Fulcra CLI access to upload, download, ingest, archive, and log third-party exports. Review any automated polling/fetching setup carefully, avoid unnecessary sensitive exports, and require confirmation before enrichment, deletion, or reprocessing workflows that could expose or mutate personal data.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
Findings (2)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The documentation instructs the agent to upload and manage user files in a remote Fulcra File Store, but it does not clearly warn that this transfers potentially sensitive third-party export data off the local environment. In an autonomous ingestion skill, that omission can cause users or downstream agents to send personal data (viewing history, purchases, health/habit exports) to remote storage without informed consent or privacy review.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill directs the agent to persist and synchronize an internal `source_map.md` registry containing source identifiers, annotation metadata, archive paths, and notes across runs, but it does not require any user notice, consent, or minimization of what gets stored. In this ingestion context, that file can accumulate sensitive operational metadata about third-party exports and historical processing, creating privacy, retention, and cross-run information exposure risks if users are unaware that such state is being maintained.

Static analysis

No suspicious patterns detected.