skills/xenodium/emacs-skills/select/Gen Agent Trust Hub

select

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to invoke emacsclient --eval to communicate with a running Emacs instance and execute Lisp commands.
  • [DYNAMIC_EXECUTION]: The skill dynamically constructs Emacs Lisp expressions by interpolating file paths and line numbers into a code template, which is then executed at runtime via emacsclient.
  • [INDIRECT_PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it reads parameters from untrusted inputs.
  • Ingestion points: File paths and line numbers are extracted from the most recent interaction context (specified in SKILL.md).
  • Boundary markers: Missing; there are no clear delimiters or validation instructions to separate safe arguments from malicious inputs.
  • Capability inventory: The skill executes shell commands via the Bash tool to run emacsclient.
  • Sanitization: Absent; file paths are inserted directly into the Emacs Lisp code block, which could lead to arbitrary Emacs Lisp execution if a file path contains escaped quotes or malicious code structures.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 02:17 AM
Security Audit — agent-trust-hub — select