tushare-cli-rust

Pass

Audited by Gen Agent Trust Hub on Mar 2, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONCREDENTIALS_UNSAFE
Full Analysis
  • [PROMPT_INJECTION]: Aggressive steering instructions in the YAML frontmatter mandate the use of this skill for any query related to Chinese financial markets, potentially overriding the agent's native intent-matching logic.
  • [COMMAND_EXECUTION]: The skill relies on the execution of a binary at ~/.claude/skills/tushare-cli/scripts/tushare and recommends using shell utilities like awk and tail to perform technical analysis calculations via command-line pipes.
  • [CREDENTIALS_UNSAFE]: The skill provides user instructions for setting the TUSHARE_TOKEN environment variable. While standard for API interaction, this requires handling sensitive credentials within the agent's environment.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its processing of untrusted external content.
  • Ingestion points: Interfaces such as news, anns_d (announcements), and research_report retrieve long-form text from external market sources.
  • Boundary markers: Absent. The instructions do not define delimiters or provide warnings to the agent to ignore instructions embedded within the retrieved data.
  • Capability inventory: The agent can execute subprocesses via the provided binary and standard system shell utilities.
  • Sanitization: Absent. There is no specified logic for sanitizing or escaping content retrieved from the Tushare API before it is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 2, 2026, 06:06 AM