github-issues
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [DATA_EXFILTRATION]: The skill reads sensitive credential files including
~/.git-credentialsand~/.hermes/.envto retrieve authentication tokens for GitHub interactions. - [EXTERNAL_DOWNLOADS]: The skill interacts with GitHub's official API (
api.github.com) usingcurlto perform issue management tasks. - [COMMAND_EXECUTION]: The skill executes shell commands such as
gh,git, andcurlto manage repositories, authenticate, and perform API operations. - [DYNAMIC_EXECUTION]: The skill employs
python3 -cone-liners to dynamically parse, filter, and format JSON data retrieved from GitHub API responses. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from GitHub issues (titles, bodies, and labels) which could contain malicious instructions designed to influence the agent.
- Ingestion points:
SKILL.md(Viewing Issues and Issue Triage Workflow sections). - Boundary markers: Absent; issue content is displayed without delimiters or specific instructions to ignore embedded commands.
- Capability inventory: File system access, network operations via
curl, and shell command execution. - Sanitization: Absent; content from issue titles and bodies is printed directly after JSON parsing.
Audit Metadata