clojure
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and modify user-provided Clojure source files and build configurations, creating a surface for potential indirect prompt injection through malicious code comments or strings. \n
- Ingestion points: Clojure source files (.clj, .cljs, .cljc) and project configuration files (deps.edn, shadow-cljs.edn, project.clj). \n
- Boundary markers: The skill does not provide specific instructions to the agent to isolate instructions embedded in data files or to treat code content as untrusted. \n
- Capability inventory: The skill utilizes CLI tools and REPL evaluation to process and run code. \n
- Sanitization: No specific sanitization or validation logic is defined for data read from the local filesystem.\n- [COMMAND_EXECUTION]: The instructions direct the agent to interact with system shells to manage project dependencies and execution. \n
- Evidence: Mentions usage of clj, clojure, npm install, and npx commands for build and dependency management.\n- [EXTERNAL_DOWNLOADS]: The documentation references several external educational and community resources. \n
- Evidence: Provides links to clojurescript.org, cljs.info, and technical blog posts at blog.janetacarr.com.
Audit Metadata