todoist

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: CRITICALREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [REMOTE_CODE_EXECUTION] (CRITICAL): The skill documentation explicitly instructs users to execute code directly from a remote URL using a piped shell command: curl -sSL https://canifi.com/skills/todoist/install.sh | bash. This allows the external site to execute arbitrary code on the user machine without prior inspection.
  • [EXTERNAL_DOWNLOADS] (HIGH): The skill downloads resources from canifi.com, which is not a recognized trusted source (e.g., official GitHub organizations of major AI companies), increasing the risk of malicious payload delivery.
  • [CREDENTIALS_UNSAFE] (MEDIUM): The skill documentation encourages users to store sensitive credentials like SERVICE_PASSWORD and TODOIST_EMAIL in an environment manager (canifi-env) that is installed via an untrusted script.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill processes external data from Todoist tasks and comments, creating a surface for indirect prompt injection.
  • Ingestion points: Todoist task titles, descriptions, and comments (SKILL.md).
  • Boundary markers: Absent; no instructions are provided to the agent to treat task content as untrusted data.
  • Capability inventory: Task management and browser interaction via Playwright MCP (SKILL.md).
  • Sanitization: Absent; the skill does not specify any sanitization or validation of the ingested task data.
Recommendations
  • HIGH: Downloads and executes remote code from: https://canifi.com/skills/todoist/install.sh, https://canifi.com/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
CRITICAL
Analyzed
Feb 17, 2026, 05:31 PM
Security Audit — agent-trust-hub — todoist