speckit-require
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION] (HIGH): The skill instructs the agent to read markdown files (e.g.,
.specify/templates/commands/specify.md) and execute the scripts found within them. It also executes.specify/scripts/bash/check-prerequisites.sh. These files are part of a repository (GitHub Spec Kit) that is not in the trusted source list, leading to the execution of untrusted code. - [COMMAND_EXECUTION] (MEDIUM): The skill utilizes the
rg(ripgrep) command to search through local specification files, which demonstrates direct shell interaction and command-line capability. - [PROMPT_INJECTION] (HIGH): The skill has a significant indirect prompt injection surface. It ingests untrusted data from
spec.mdandplan.mdfiles to drive its core workflow. - Ingestion points: Processes external specification files and user input.
- Boundary markers: Absent in the prompt logic.
- Capability inventory: Can execute shell scripts, run
rg, and perform file system write operations. - Sanitization: None detected. Malicious instructions embedded in the specification files could be interpreted as commands by the agent.
Recommendations
- AI detected serious security threats
Audit Metadata