sync-jira-bug
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from local markdown files to populate Jira issue descriptions, creating an attack surface for indirect prompt injection. Ingestion points include markdown bug reports in the repository's docs/ directory. Capability inventory includes network operations via the Jira API, shell execution of development tools, and local filesystem writes. No specific boundary markers or 'ignore' instructions are visible to separate untrusted content from the internal execution context.
- [COMMAND_EXECUTION]: The skill interacts with the local environment using standard developer tools. It executes git via execFileSync and execSync for detection and history tasks. It also uses the gh CLI for GitHub interactions and awk for text processing and environment resolution.
- [EXTERNAL_DOWNLOADS]: The skill connects to well-known cloud services to facilitate synchronization. It manages network requests to Atlassian Jira and Bitbucket platforms for issue tracking and source document link generation.
- [CREDENTIALS_UNSAFE]: The skill handles authentication tokens for Jira and Bitbucket but follows strong security practices. It utilizes environment variables for secret management and implements a recursive redaction system in references/defer-mutation.js that masks secret values in journals and logs.
- [DYNAMIC_EXECUTION]: The skill employs localized dynamic execution to ensure reliable configuration parsing. references/read-config.sh contains an embedded Python script executed via python -c to provide secure YAML parsing and enforce strict rules like duplicate-key rejection, preventing security escalations from malformed configuration files.
Audit Metadata