upgrade-plugin

Pass

Audited by Gen Agent Trust Hub on Mar 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Retrieves latest release notes and changelogs from Anthropic's official GitHub repository (github.com/anthropics/claude-code).
  • [EXTERNAL_DOWNLOADS]: Fetches current documentation and model capability updates from Anthropic's official site (docs.anthropic.com).
  • [COMMAND_EXECUTION]: Utilizes standard shell utilities (ls, find, cat, wc, head) to inventory and audit local plugin files within the user's plugin cache and current working directory.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from external URLs and other installed plugins to derive upgrade patterns. ● Ingestion points: Content retrieved via WebFetch and WebSearch from Anthropic domains; local file reads of other plugins via Glob and Read. ● Boundary markers: The instructions do not define specific delimiters or 'ignore' instructions for the ingested content. ● Capability inventory: The skill has the capability to write to local files (Phase 4) and execute shell commands for verification. ● Sanitization: There is no explicit sanitization or validation logic for the ingested markdown and text content before it is used to influence file modifications.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 15, 2026, 01:57 AM