The Agent Skills Directory

[COMMAND_EXECUTION]: The script bin/hunter.py executes the gh (GitHub CLI) tool via subprocess.run to perform repository discovery and issue listing.
The execution uses an argument list (['gh'] + args) rather than a shell string, which prevents standard shell command injection.
This functionality is central to the skill's primary purpose of automating GitHub searches.
[INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes untrusted data from the GitHub API, including repository descriptions and issue titles, which could contain malicious instructions designed to influence agent behavior.
Ingestion points: External data is ingested in bin/hunter.py through calls to gh api and gh issue list.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the retrieved GitHub content as untrusted or to ignore instructions embedded within it.
Capability inventory: The skill has the ability to execute shell commands via the gh tool and can access the internet to fetch repository data.
Sanitization: The script does not perform sanitization, escaping, or validation on the text returned by the GitHub API before outputting it to the agent's context.

oss-hunter