teach-me
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses subprocesses to execute
git diff,git show, andgh pr diffto retrieve code context for generating lessons on specific changes. - [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests untrusted data from multiple sources and interpolates them into generated artifacts.
- Ingestion points: Data enters the context via local codebase files (SKILL.md), git diff/show output (references/code-change.md), GitHub PR diffs, and external research results from supplementary skills.
- Boundary markers: The instructions do not define specific delimiters or boundary markers to isolate untrusted data within the prompt or output artifacts.
- Capability inventory: The skill has the capability to read from and write to the local file system (saving artifacts to
.teach/) and execute git-related shell commands. - Sanitization: There are no explicit sanitization or escaping instructions provided for handling untrusted code or research data before it is rendered into HTML or Markdown artifacts.
Audit Metadata