ghidra-mcp-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from binaries (e.g., decompilation, strings, and imports). A maliciously crafted binary could contain strings or code patterns intended to mislead the agent's analysis or trick it into performing unintended actions within Ghidra.
- Ingestion points: The agent receives untrusted data through tools like
list_strings,decompile_function,list_imports, andlist_data_itemsfromreferences/workflow.mdandSKILL.md. - Boundary markers: The instructions lack specific boundary markers or "ignore embedded instructions" warnings for the data retrieved from the binary.
- Capability inventory: The skill possesses write capabilities that can modify the state of the Ghidra project, including
rename_function,rename_data,set_decompiler_comment,set_function_prototype, andset_local_variable_type. - Sanitization: There is no evidence of sanitization or validation of the content retrieved from the binary before it is processed by the agent.
Audit Metadata