xy-archive
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including
basename $(pwd)for project naming andmkdir -pfor directory setup in the archive storage process. - [DYNAMIC_EXECUTION]: The instruction in
SKILL.mdspecifies the execution of a Python command stringpython3 -c 'import datetime as d; print(d.datetime.now().astimezone().isoformat("T","seconds"))'to generate metadata timestamps. - [INDIRECT_PROMPT_INJECTION]: The skill creates a vulnerability surface by ingesting untrusted user conclusions and titles which are stored for later use by other agents.
- Ingestion points: User-supplied conclusion text and titles in
SKILL.md. - Boundary markers: Uses YAML frontmatter and specific Markdown headers as delimiters.
- Capability inventory: Local file system writes, directory creation, and shell command execution.
- Sanitization: No explicit content sanitization or escaping mechanisms are specified before writing data to files.
Audit Metadata