setup-project-context

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill facilitates the injection of content from external files (e.g., CLAUDE.md, AGENTS.md, and .claude/rules/*.md) into the agent's session context. This creates a surface for indirect prompt injection if the agent interacts with untrusted repositories containing these files.
  • Ingestion points: The skill configures the agent to read .claude/project-context.json which points to external repository paths.
  • Boundary markers: The instructions do not specify any delimiters or safety warnings for the injected content.
  • Capability inventory: The environment includes file read/write access and the ability to execute shell commands.
  • Sanitization: No evidence of sanitization or filtering for the external content is provided.
  • [COMMAND_EXECUTION]: The skill configures postToolFormatCommands, which are shell commands executed sequentially in the target project root after Edit or Write operations. While intended for formatting, this provides a mechanism for arbitrary command execution triggered by file modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 12:04 PM
Security Audit — agent-trust-hub — setup-project-context