context-optimizer

Pass

Audited by Gen Agent Trust Hub on Oct 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of the Bash tool to perform system audits, environment variable checks, and process management.
  • It includes commands to find and kill zombie processes (kill, pkill) and remove session data or state files (rm).
  • It audits sensitive configuration files such as ~/.claude/settings.json and .env to identify optimization opportunities.
  • It performs broad filesystem searches, such as find / -maxdepth 6 -name "CLAUDE.md", to identify configuration cascades.
  • [EXTERNAL_DOWNLOADS]: The skill encourages the installation of external software to support its optimization workflow.
  • It recommends installing the ccusage package via npm or cloning a fork directly from the author's GitHub repository (LEON-gittech/ccusage) for cost analysis.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes various files within the user's workspace, which could potentially contain malicious instructions intended to influence the agent's behavior during the audit.
  • Ingestion points: Audits content from CLAUDE.md files (across multiple directory levels), project-specific .claude/settings.json files, and plugin hooks.json files.
  • Boundary markers: None detected; the skill does not explicitly instruct the agent to use delimiters or ignore instructions embedded within the files it reads.
  • Capability inventory: The skill utilizes Read, Glob, Grep, Bash, and Edit tools, providing a wide range of actions that could be triggered by injected instructions.
  • Sanitization: There are no described methods for sanitizing or filtering the content processed from project files before it is evaluated by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 1, 2026, 01:21 PM