fetching-work-item

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references official documentation from well-known and trusted services, including Atlassian, GitHub, and Anthropic, to provide the agent with just-in-time information on API syntax and best practices.
  • [COMMAND_EXECUTION]: Utilizes the standard gh CLI and dedicated Jira MCP tools to perform read-only queries. These operations are restricted to retrieving work-item data as part of the skill's primary function.
  • [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data from external sources and implements appropriate safeguards.
  • Ingestion points: External content from GitHub issues and Jira tickets (handled in subagents/work-item-retriever.md).
  • Boundary markers: Employs a structured Markdown template (references/github-snapshot-template.md, references/jira-snapshot-template.md) with a specific preamble and metadata tables.
  • Capability inventory: Limited to writing a single local file in the docs/ directory per work item.
  • Sanitization: Includes a robust 'Heading rewrite' rule in references/retrieval-playbook.md that transforms external Markdown headings into bold labels, preventing external content from mimicking the snapshot's reserved section headers.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 06:40 AM
Security Audit — agent-trust-hub — fetching-work-item