backlog
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes
docs/devdocs/backlog.md, a file that can be populated with content from other agents or users, creating a surface for indirect prompt injection. - Ingestion points:
docs/devdocs/backlog.md(read viaReadandGreptools). - Boundary markers: The skill uses YAML frontmatter blocks and Markdown headers to segment entries, providing clear delimiters between distinct task items.
- Capability inventory: The skill utilizes
WriteandAskUserQuestiontools, allowing it to modify files based on instructions interpreted from the backlog. - Sanitization: The skill does not explicitly sanitize the text content within backlog entries before processing, relying on the structural delimiters and platform guardrails.
Audit Metadata