The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute standard diagnostic commands such as llvm-dwarfdump, readelf, and grep to extract and filter debug information from binary files.
[EXTERNAL_DOWNLOADS]: References authoritative documentation and source code from trusted locations including the official DWARF standard website (dwarfstd.org) and well-known GitHub repositories for projects like LLVM, libdwarf, and pyelftools.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted binary data that could contain maliciously crafted DWARF metadata designed to influence agent behavior.
Ingestion points: The skill reads and parses user-provided binary files via tools like dwarfdump and readelf (documented in SKILL.md and reference/dwarfdump.md).
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands in the parsed output are provided.
Capability inventory: The skill has access to Bash (for command execution) and WebSearch, which could be exploited if the agent obeys instructions found within debug metadata.
Sanitization: There is no evidence of sanitization or validation of the DWARF content before it is processed by the agent.

dwarf-expert