archify
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill implements legitimate diagramming functionality. It includes a dedicated security module (
renderers/shared/output-path.mjs) specifically designed to prevent path traversal and unauthorized file overwrites by validating that output paths remain within the current working directory. - [SAFE]: All user-provided data is strictly validated against JSON schemas (
schemas/*.schema.json) withadditionalProperties: falseto prevent unexpected data ingestion or schema-based attacks. - [SAFE]: The skill uses comprehensive escaping mechanisms (
escfunction and Unicode JSON escaping) for all string interpolation to prevent Cross-Site Scripting (XSS) in the generated HTML artifacts. - [COMMAND_EXECUTION]: The skill uses
spawnSyncto executegitcommands for its 'repository evidence' feature. This execution is limited to specific Git commands (rev-parse,remote,cat-file,show) used to verify code existence and is guarded by path validation and revision syntax checks. - [SAFE]: The local preview server uses standard
http.createServerbound to127.0.0.1and is only active during the user-initiated authoring loop.
Audit Metadata