speckit-extension

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for the agent to download and install extensions from user-specified remote URLs (--from <url>) and to add custom HTTPS extension catalogs.
  • [COMMAND_EXECUTION]: All operations rely on executing the specify CLI tool in the local environment.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill ingests external content from remote URLs and catalogs, placing them into the project's skills directory (.github/skills/).
  • Boundary markers: None. There are no instructions for the agent to validate or delimit instructions contained within downloaded extensions.
  • Capability inventory: The skill uses the specify CLI to write files to the filesystem and manage extension lifecycles.
  • Sanitization: The --install-allowed flag for catalogs acts as a basic manual permission gate for new sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 11:49 PM
Security Audit — agent-trust-hub — speckit-extension