dsh-archive-agent-notes
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local scripts via pnpm for tasks such as documentation synchronization (doc-sync), hash verification (verify-archived-agent-notes), and linting. It also utilizes git commands to verify file changes.
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to read, audit, and semantically judge 'Agent Notes' (markdown files) to determine if they should be archived, kept active, or deleted. This creates a surface where instructions embedded in those files could influence agent behavior.
- Ingestion points: External markdown files including Agent Note triplets (foo.md, foo.zh.md) and configuration files like ../../notes/README.md and ../../notes/archived/AGENTS.md.
- Boundary markers: No explicit delimiters, XML tags, or 'ignore instructions' warnings are provided to separate the data being processed from the agent's core instructions.
- Capability inventory: The skill allows the agent to move files across directories, modify frontmatter metadata, and execute shell scripts (pnpm, git).
- Sanitization: There is no evidence of sanitization, escaping, or validation of the note content before it is read and evaluated by the agent.
Audit Metadata