speckit-baseline
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The workflow involves executing a local bash script located at
.specify/scripts/bash/create-new-feature.sh. Arguments for this script, such as a feature 'short name', are dynamically generated by the AI after analyzing repository source code. The skill includes specific instructions for the agent to escape single quotes when constructing this command to mitigate potential injection.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code to generate feature specifications, which could be exploited to influence agent behavior.\n - Ingestion points: Workflow step 2 reads source files based on user-provided paths or glob patterns.\n
- Boundary markers: Absent; the skill does not define specific delimiters or instructions to ignore embedded commands within the analyzed code.\n
- Capability inventory: The agent can execute a specific bash script (
create-new-feature.sh) and write files to thespecs/directory.\n - Sanitization: The skill provides manual escaping instructions for the agent when passing arguments to the shell script, but does not specify automated sanitization or filtering of the content extracted from analyzed source files.
Audit Metadata