kstack-generate-principles
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from an Obsidian vault to derive new principles and playbooks for the agent.
- Ingestion points: Reads various MOC (Map of Content) files,
index.md, and searchable body text from the user's Obsidian vault. - Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the vault content.
- Capability inventory: The skill can write to the local filesystem to update repository files (
SKILL.md,playbooks/*.md) via thewriting-for-agentsskill. - Sanitization: There is no evidence of sanitization or filtering to prevent vault-based instructions from overriding agent logic.
- [COMMAND_EXECUTION]: The skill modifies repository files during the application phase.
- It uses the
writing-for-agentsskill to perform file edits on approved principles and playbooks within thekriscard-stackrepository. - [EXTERNAL_DOWNLOADS]: The skill references external content for benchmarking quality.
- It points to a public repository on GitHub (
github.com/cursor/plugins) to establish standards for its generated artifacts. This is a reference to a well-known service for documentation purposes.
Audit Metadata