skills/konglong87/superpm/write-prd/Gen Agent Trust Hub

write-prd

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill logic involves reading existing documents from the docs/01-需求调研/ directory to determine the execution flow (e.g., whether to skip certain steps). If these project files contain untrusted or adversarial content, it could manipulate the agent's behavior during the document generation process or during tool usage.\n
  • Ingestion points: The skill accesses the local filesystem to read existing documents in the docs/01-需求调研/ folder via the Read tool.\n
  • Boundary markers: The instructions lack explicit delimiters or instructions for the agent to treat the content of read files as untrusted data or to ignore any embedded directives.\n
  • Capability inventory: The execution environment provides access to sensitive tools such as Bash, Write, and WebSearch, as well as the ability to invoke other skills.\n
  • Sanitization: No evidence was found of content validation or sanitization being applied to the data retrieved from the local files before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 02:54 AM
Security Audit — agent-trust-hub — write-prd