shared-world

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Deno scripts to initialize and manage the world-bible directory structure. These scripts, including init-world.ts, add-entry.ts, and build-index.ts, perform file system read and write operations based on agent-provided arguments.- [EXTERNAL_DOWNLOADS]: Scripts import standard modules from the official Deno registry, which is a well-known and trusted source for utility libraries. Specifically, build-index.ts and check-conflicts.ts import from https://deno.land/std@0.208.0/fs/walk.ts.- [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection as it processes user-controlled text files that could influence agent behavior.
  • Ingestion points: The check-conflicts.ts and build-index.ts scripts read and parse user-provided markdown entries in the world-bible directory.
  • Boundary markers: While data is structured with markdown headers, the skill lacks explicit delimiters or instructions to prevent the agent from interpreting embedded malicious text in full-text fields like 'Summary' or 'Description'.
  • Capability inventory: The agent has the capability to write to the local file system and execute shell commands via Deno.
  • Sanitization: Filenames are sanitized through a slugify regex, but the actual content of the markdown files is not validated or sanitized against command-like natural language patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 05:35 PM