chengfeng-report-bug

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes gh (GitHub CLI) to interact with GitHub repositories. This is the core functionality and is used securely with argument quoting and validation.
  • [DATA_EXFILTRATION]: The skill provides a robust data sanitization mechanism in scripts/report-bug.cjs. It automatically redacts common sensitive patterns such as API keys, GitHub tokens, JWTs, and local file paths before generating a bug report draft. It also enforces a mandatory user confirmation step, where the user must review the de-identified content before it is uploaded to GitHub.
  • [REMOTE_CODE_EXECUTION]: The skill downloads and executes node scripts/ensure-runtime.cjs for diagnostics. This is a local script within the skill's own package, used for environment checks. The execution of remote content is limited to interaction with well-known GitHub repositories through the official GitHub CLI.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided bug descriptions. To mitigate risks, it uses structured JSON input and a strict multi-step process: diagnostic collection, sanitization, draft preview, and explicit user confirmation. These boundaries ensure that untrusted user input cannot manipulate the agent's actions or the final submission without review.
  • [PRIVILEGE_ESCALATION]: The skill uses umask 077 when writing temporary files to ensure they are only readable by the current process, adhering to the principle of least privilege.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 01:08 AM
Security Audit — agent-trust-hub — chengfeng-report-bug