writing-skills
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The utility
render-graphs.jsexecutes the system'sdotcommand usingchild_process.execSyncto generate SVG diagrams. - [EXTERNAL_DOWNLOADS]: Installation instructions for the
graphvizdependency refer to trusted system package managers (brewandapt). - [INDIRECT_PROMPT_INJECTION]: The skill methodology processes markdown content to extract Graphviz code. Ingestion points:
render-graphs.jsreadsSKILL.md. Boundary markers: markdown triple backticks. Capability inventory:execSyncand file system access. Sanitization: none. - [PROMPT_INJECTION]: The skill employs authoritative language and psychological principles ('YOU MUST', 'No exceptions') to enforce the TDD workflow, which is a functional process requirement rather than an attempt to circumvent safety filters.
Audit Metadata