series-continuity
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input, specifically the project title and synopsis, which is interpolated into shell commands. This represents a vulnerability surface where malicious input could attempt to execute arbitrary commands if not properly handled.
- Ingestion points: The skill instructs the agent to ask the user for a Title and synopsis in
SKILL.md. - Boundary markers: The instructions mandate wrapping user-supplied values in single quotes to delimit them from the rest of the command.
- Capability inventory: The skill utilizes shell execution for
story init,story reindex, and other project management tasks. - Sanitization: The skill includes explicit instructions for the agent on how to escape single quotes according to the specific shell environment (POSIX vs. PowerShell) to prevent command breakout.
- [DYNAMIC_EXECUTION]: The skill directs the agent to execute JavaScript files using computed relative paths and user-defined checkout locations.
- Evidence:
SKILL.mdspecifies a fallback tonode ../story-maintenance/scripts/story.jsand allows execution from a user-named checkout path atnode <checkout>/bin/story.js. - [COMMAND_EXECUTION]: The skill relies on executing various shell commands to perform project initialization, reindexing, and continuity checks.
- Evidence: Documentation and instructions for running
story init,story reindex,story wordcount,story check, andstory seriesare central to the skill's operation.
Audit Metadata