Skills
Navigate
skills/jeremylongshore/tons-of-skills-marketplace/agent-context-loader/Gen Agent Trust Hub

agent-context-loader

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill discovers and merges instructions from external files into the agent session.
  • Ingestion points: Discovery of AGENTS.md files across the directory tree and ancestors (SKILL.md).
  • Boundary markers: No explicit delimiters or boundary instructions are mentioned for the merged content.
  • Capability inventory: The skill utilizes Bash, Grep, Read, Write, and Edit tools (SKILL.md).
  • Sanitization: The instructions do not define sanitization logic for the ingested instruction blocks.
  • [COMMAND_EXECUTION]: The skill uses Bash for filesystem discovery and includes a directory-change hook script scripts/check-agents-md.sh that checks for file existence and outputs text to the terminal.
  • [EXTERNAL_DOWNLOADS]: The skill links to official documentation (docs.anthropic.com) and the author's personal and business domains (jeremylongshore.com, intentsolutions.io, tonsofskills.com). These resources are recognized as legitimate and trace back to the author's infrastructure.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:41 AM