ot-handoff
Pass
Audited by Gen Agent Trust Hub on Nov 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local Bash script at
../ot-init/scripts/on-track-index.sh. This script is part of a sibling directory structure, presumably within the same vendor's toolset, and is used to update the project index. It is described as having no network access. - [INDIRECT_PROMPT_INJECTION]: The skill creates a persistent storage of conversation state that includes verbatim user instructions. This poses a risk where malicious instructions in the current session could be passed to and executed by a future agent picking up the handoff.
- Ingestion points: Conversation history and active user prompts.
- Boundary markers: The skill uses markdown blockquotes in the
assets/handoff.mdtemplate to delimit instructions, which provides a weak boundary against adversarial content. - Capability inventory: The skill has the ability to write to the filesystem and execute local scripts.
- Sanitization: The instructions explicitly forbid paraphrasing user input (Step 2 in
SKILL.md), which intentionally bypasses normalization or sanitization of potentially malicious instructions.
Audit Metadata