Back to skill

Security audit

Prompt Compression

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be a local prompt-compression tool, but its default rules can broadly rewrite agent instruction files in ways that may change their meaning.

Review this carefully before installing. It does not show evidence of exfiltration or hidden execution, but its defaults are risky for system prompts and skill files. Only use it on version-controlled files, inspect the full diff before running an agent with compressed prompts, and avoid applying it to safety, credential, authorization, or tool-permission instructions unless you have manually verified that meaning is preserved.

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
assets/trinity-compress.config.json:6
Finding

Context-Insensitive Rewriting Can Weaken Security-Sensitive Agent Instructions

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

This configuration targets prompt and skill files, then removes entire phrases such as cautionary or emphasis-oriented language without any protected-term list or safety-aware exclusions. In prompt-governing files, broad phrase deletion can silently strip warnings, constraints, or risk qualifiers and thereby weaken downstream safety behavior or change instruction meaning.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The instruction-compression rules rewrite normative prompt language globally across system and skill prompt targets, including transformations around obligations and prohibitions such as replacing or deleting phrases like 'You should', 'You need to', and contractions of 'do not'. Because these files define agent behavior, overly broad rewriting can materially alter force, scope, or clarity of security instructions and create prompt-meaning drift that weakens safeguards.

Content

No source excerpt is available for this finding.