ai-brag-document
Pass
Audited by Gen Agent Trust Hub on Jul 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell command templates with placeholders that could be susceptible to injection.
- Evidence: In Pass 5 of the discovery phase, the skill defines
AUTHOR_PATTERN='<name|email|github-username|known-alias>'and passes it directly torg. If the agent populates these placeholders with unsanitized user-provided aliases containing shell metacharacters, it could result in arbitrary command execution. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting and processing untrusted data from local files.
- Ingestion points: The skill reads content from all Markdown files within the user's engineering vault (e.g.,
cat "$ENGINEERING_PATH/<path>.md") to identify accomplishments. - Boundary markers: The skill does not implement boundary markers or instructions to ignore embedded commands within the processed notes.
- Capability inventory: The agent has access to file system read/write operations and local Git repository management (
git add,git commit). - Sanitization: There is no evidence of sanitization or validation of the content read from the Markdown notes before it is used to generate the final document or influence the Git workflow.
Audit Metadata