tb-update-task-note
Warn
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local script
bin/tb-api.mjsusing the Node.js runtime to perform API operations. - [COMMAND_EXECUTION]: User-provided notes are inserted directly into a shell command string (
--note "<备注>"). This instruction is vulnerable to command injection if the input contains shell control characters like semicolons or backticks, allowing for potential unauthorized command execution. - [DATA_EXFILTRATION]: The skill reads project and task data from a local file named
.teambition.md. While necessary for the skill's functionality, this involves access to local configuration data. - [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by reading data from
.teambition.mdand using it to guide agent actions. 1. Ingestion points:.teambition.md. 2. Boundary markers: Absent. 3. Capability inventory:node bin/tb-api.mjscommand execution. 4. Sanitization: Absent.
Audit Metadata