Dex Task Tracking
Analysis
This is a simple local task-tracking instruction skill, with minor notes that it persists task details on disk and assumes a trusted dex command is available.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
dex create -d "Description" --context "Background, goal, done-when"
The runtime instructions rely on a `dex` command, while the provided artifacts include no install spec or required-binary declaration. This is a provenance/completeness note rather than evidence of unsafe behavior.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
Tasks stored as JSON files in .dex/tasks/.
The skill deliberately persists task descriptions, context, and results across sessions in local files, which can include project details or follow-up instructions.
