context-surfing

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data from multiple project files and local session artifacts, creating an indirect prompt injection surface.
  • Ingestion points: Reads CLAUDE.md, AGENTS.md, README.md, and all root-level .md files, plus handoff files in the .context-surfing/ directory.
  • Boundary markers: Absent; instructions and context from these files are directly incorporated into the agent's operation.
  • Capability inventory: The skill writes handoff files to the filesystem and executes shell scripts and CLI commands.
  • Sanitization: No sanitization or validation of content from external files is performed.
  • [COMMAND_EXECUTION]: The skill uses a shell script (scripts/handoff-checker.sh) as a hook in agent settings and executes entire status to interface with the Entire CLI for session tracking.
  • [EXTERNAL_DOWNLOADS]: The skill is installed from a remote source via npx skills add pskoett/pskoett-ai-skills/skills/context-surfing, which is a repository maintained by the author.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 03:06 PM