update-context

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local commands using 'ls' to check for directory existence and 'uv run python' to execute a synchronization script. These commands target specific local paths within the user home directory (~/.claude/.context/) and the skill root directory.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection. It is designed to extract instructions, rules, and facts directly from untrusted conversation history and persists them to long-term storage.
  • Ingestion points: Conversation history (SKILL.md)
  • Boundary markers: None; the skill lacks delimiters or instructions to ignore embedded commands within the conversation
  • Capability inventory: Subprocess execution via 'uv run python' and persistent file-write operations (SKILL.md)
  • Sanitization: No validation or sanitization of the extracted data is performed before it is written to the file system.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 01:56 AM