xgent-init
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local validation script (
scripts/check-docs.mjs) using the Node.js runtime. This script performs structural analysis on the generated Markdown files to prevent common errors like residual placeholders or malformed headers. This execution is scoped to the local environment and the skill's own assets. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect injection surface by ingesting untrusted data from local repository files (e.g.,
app.manifest.json,package.json,Dockerfile) to fill documentation templates. - Ingestion points: Data is read from the root directory or specific deployment paths of the target repository.
- Boundary markers: The skill uses explicit token delimiters (e.g.,
<APP_KEY>) and mandates their removal before completion. - Capability inventory: File read/write operations and local script execution.
- Sanitization: Mitigation is implemented via "Red Line" rules prohibiting data fabrication and a mandatory validation script that fails if structural anomalies or placeholder residues are detected.
Audit Metadata