ptlam-code-style-typescript

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill instructs the agent to ingest and prioritize directives from repository-controlled files such as AGENTS.md and local source code, creating an indirect prompt injection surface.\n
  • Ingestion points: SKILL.md (Before the first edit) and skills/ptlam-code-style/SKILL.md (Who decides what table).\n
  • Boundary markers: Absent; no instructions are provided to the agent to disregard potential injections within these data-carrying files.\n
  • Capability inventory: Shell execution via npm, tsc, vitest, and biome.\n
  • Sanitization: Absent.\n- [COMMAND_EXECUTION]: The skill establishes a standard local development loop that executes shell commands to perform type checking, linting, and testing within the project environment.\n- [EXTERNAL_DOWNLOADS]: The instructions recommend installing standard, reputable packages such as typescript, vitest, zod, and pino from the npm registry.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 10:29 AM
Security Audit — agent-trust-hub — ptlam-code-style-typescript