pm-pool
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill's preamble attempts to execute a local script named
check-update.sh(bash "$(dirname "${BASH_SOURCE[0]}")"/check-update.sh). Since this script is not part of the provided analysis, its exact behavior is unverifiable, though it follows a common vendor pattern for lifecycle management.- [PROMPT_INJECTION]: The skill facilitates an indirect prompt injection surface by collecting and storing unsanitized user input into a persistent project document that is later read back into the agent's context. - Ingestion points: User-provided 'Requirement Name' and 'Description' in the 'Add New Requirement' operation.
- Boundary markers: Absent; user content is interpolated directly into the Markdown table and detail sections without delimiters or instructions to the agent to ignore embedded commands.
- Capability inventory: The skill has access to
Bash,Read,Write, andEdittools, which could be abused if malicious instructions are embedded in the requirement descriptions. - Sanitization: No sanitization, validation, or escaping is performed on the input before it is written to
docs/01-需求调研/需求池管理表.md.
Audit Metadata