tracekit-auth

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local management script ./scripts/run-tracekit-auth.sh to perform status checks, user registration, and code verification.\n- [DATA_EXFILTRATION]: User-provided authentication data, including email addresses and 6-digit verification codes, is transmitted to the official vendor endpoint https://app.tracekit.dev. This is consistent with the skill's stated purpose of facilitating TraceKit authentication.\n- [CREDENTIALS_UNSAFE]: The skill reads from and writes to the ~/.tracekitconfig file and interacts with the TRACEKIT_API_KEY environment variable to maintain persistent authentication states.\n- [PROMPT_INJECTION]: The skill ingests untrusted user data (email and verification code) and passes them as arguments to a shell script, which represents an indirect prompt injection surface.\n
  • Ingestion points: User email address and 6-digit verification code input in SKILL.md workflow.\n
  • Boundary markers: Absent; user inputs are interpolated directly into command arguments without explicit delimiters.\n
  • Capability inventory: Execution of local subprocesses via ./scripts/run-tracekit-auth.sh.\n
  • Sanitization: No sanitization or validation steps for the email or code inputs are specified in the markdown instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 10:15 AM