directions-lab
Warn
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: MEDIUMDYNAMIC_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [DYNAMIC_EXECUTION]: The skill uses dynamic module imports (
import()) on paths constructed at runtime. This behavior is found inscripts/lib/capabilities.mjs, which loads a feature index script from the project root, and inscripts/lib/render.mjs, which dynamically loads various file templates during the scaffolding process. - [COMMAND_EXECUTION]: The scaffolding logic in
scripts/scaffold.mjsandscripts/lib/verify.mjsexecutes system commands viachild_process.spawnSyncandexecFileSync. These are used to verify the generated lab by runningvitest,vite build, andpnpm typecheck. The skill also generates shell command configurations for user launch profiles in.claude/launch.json. - [INDIRECT_PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it ingests user input via CLI flags and interpolates it into generated source code.
- Ingestion points: Flag inputs in
scripts/scaffold.mjssuch as--slug,--title, and--surfaces. - Boundary markers: None identified in the templates where flags are interpolated.
- Capability inventory: Execution of shell commands via
spawnSyncinscripts/lib/verify.mjsand file system modifications/deletions inscripts/scaffold.mjs. - Sanitization: Slugs and IDs are validated with regular expressions, but descriptive labels and names used in templates receive less stringent filtering.
- [EXTERNAL_DOWNLOADS]: The skill instructions in
SKILL.mdadvise runningpnpm installto manage dependencies. Additionally, the capture utilityassets/templates/shot.cjs.mjsrelies on the externalplaywrightpackage and its browser binaries to generate screenshots.
Audit Metadata