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

describe

Warn

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool to run emacsclient --eval. This mechanism allows the agent to execute Lisp code within a running Emacs process, granting control over the Emacs environment.\n- [DYNAMIC_EXECUTION]: The skill uses load to dynamically incorporate the agent-skill-describe.el script into the Emacs environment at runtime and constructs Lisp expressions through string concatenation.\n- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to Lisp injection via the user-supplied query parameter.\n
  • Ingestion points: The user search term in SKILL.md.\n
  • Boundary markers: Absent. The query is placed inside a Lisp string literal without escaping instructions.\n
  • Capability inventory: Bash tool access and emacsclient --eval capability.\n
  • Sanitization: Absent. No instructions are provided to sanitize or escape input characters like quotes or parentheses.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 15, 2026, 02:18 AM
Security Audit — agent-trust-hub — describe