obsidian-log-commits
Pass
Audited by Gen Agent Trust Hub on May 7, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the GitHub CLI (
gh) to dynamically discover and query data from external repositories. - Evidence: Uses
gh repo list <username>,gh api /user/orgs, andgh repo list <org>to identify targets for commit fetching. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from commit messages.
- Ingestion points: External GitHub commit messages fetched via the CLI (SKILL.md, Section 2).
- Boundary markers: Absent. The instructions do not provide delimiters or specific warnings to ignore instructions found within the commit data.
- Capability inventory: The agent has the ability to execute shell commands (
gh) and read/write files within the Obsidian vault (SKILL.md, Section 5). - Sanitization: Absent. The skill instructs the agent to "synthesize" the information into bullets, but does not specify escaping characters or validating the content against malicious markdown or shell instructions.
Audit Metadata