Back to skill

Security audit

research-logger

Security checks for vulnerabilities and agentic risk

Overview

The skill’s research-note purpose is coherent, but its script has unsafe input handling that can turn user or web-derived text into unintended local command execution.

Review this skill before installing. Do not use it with confidential research topics, and avoid running the script on untrusted input until the sed rendering is replaced with a data-safe templating method and tag arguments are passed as a quoted array. Pin the grizzly dependency to a reviewed version if you choose to use it.

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)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/research_logger.sh:58
Finding

Untrusted Input Injected into Dynamically Constructed sed Programs

Content
View full analysis
Remediation
View remediation

T08 · Insecure Dependencies

Note
Location
SKILL.md:51
Finding

Mutable Unpinned Dependency Installation Instruction

Content
View full analysis
Remediation
View remediation
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 (4)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill documentation describes use of web search, web fetch, and GIF lookup, but it does not clearly warn users that their supplied topic and potentially derived content will be transmitted to external services. This creates a privacy and data-handling risk, especially if users provide sensitive research topics, internal project names, or confidential material under the assumption the workflow is local except for note creation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The script builds TAG_FLAGS by concatenating untrusted tag values into a shell command string and then expands that string unquoted in the grizzly invocation. Because shell metacharacters and command substitutions in user-controlled tags can be reinterpreted during expansion, an attacker can inject additional arguments or execute unintended commands when the note is created.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
89% confidence
Finding

The script sends the supplied research topic to web_search, which is a network-facing operation, but only prints a generic search status line. There is no explicit warning or documentation that the user's input may be transmitted to external search services.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
86% confidence
Finding

The gifgrep invocation appears to perform an external lookup based on the provided topic, but the script gives no explicit disclosure that user input is being shared beyond the local system. For operations that may transmit user data, the rules require some form of visible warning or documentation.

Content

No source excerpt is available for this finding.