The Agent Skills Directory

[DATA_EXFILTRATION]: The skill implements a telemetry system in its preamble and post-workflow scripts. It reads a remote endpoint and telemetry tier from ~/.superstack/config.json and uses curl to transmit usage data, including skill phase, status, and system platform information (OS and architecture), to a Convex-hosted endpoint. The skill explicitly requests user consent before activating this configuration.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through its state management mechanism. It reads from .superstack/build-context.md at the start of the workflow to retrieve prior decisions. Ingestion points: Reads .superstack/build-context.md in the SKILL.md workflow. Boundary markers: No specific delimiters or safety instructions are used to isolate content read from the context file. Capability inventory: The skill can execute shell commands via spl-token and npm, and perform network operations using curl. Sanitization: No validation or escaping is applied to the data retrieved from the build context.
[EXTERNAL_DOWNLOADS]: The skill instructs the user to install established Solana ecosystem libraries, including @metaplex-foundation/mpl-token-metadata, @metaplex-foundation/umi, and @pump-fun/pump-sdk, via the npm package manager. These are well-known tools for token metadata and launch operations.

launch-token