create-evlog-enricher
Pass
Audited by Gen Agent Trust Hub on Mar 15, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: Indirect prompt injection surface via multi-step file modification.
- Ingestion points: User-provided placeholders
{name},{Name}, and{DISPLAY}defined in the naming conventions ofSKILL.md. - Boundary markers: Absent; the instructions direct the agent to interpolate user input directly into file content without delimiters or instructions to ignore embedded commands.
- Capability inventory: The skill has the capability to modify source code (
packages/evlog/src/enrichers/index.ts) and a sibling AI agent skill (skills/review-logging-patterns/SKILL.md), which could allow for persistence of injected instructions. - Sanitization: No validation, escaping, or filtering of the placeholders is specified before they are written to disk.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for project verification.
- Evidence: Step 6 (Verification) commands
bun run buildandbun run test. These are standard operations within the intended development workflow for this repository.
Audit Metadata