Hedera Agent Kit v4 Migration
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to perform project-wide searches (
grep) and read various source files (.ts,.js) and dependency manifests (package.json). This workflow introduces a surface for indirect prompt injection, as malicious instructions hidden within the project's codebase or metadata could potentially hijack the agent's context during the migration. - Ingestion points: Local project files, including source code and
package.json, accessed through search commands inSKILL.md. - Boundary markers: Not present; the skill does not provide instructions to delimit or ignore instructions found within the audited code.
- Capability inventory: The agent is tasked with file modifications and package installations (
npm install), providing a significant capability surface. - Sanitization: No mechanisms for sanitizing or escaping the content of processed files are specified.
- [EXTERNAL_DOWNLOADS]: The skill guides the agent to install updated dependencies from official package registries.
- Evidence: Mentions migration to official packages including
@hashgraph/hedera-agent-kitand@hiero-ledger/sdk.
Audit Metadata