The Agent Skills Directory

[PERSISTENCE]: The skill provides explicit instructions for users to persist the TINYFISH_API_KEY by appending export commands to shell initialization files (~/.bashrc or ~/.zshrc) and by modifying local application settings (~/.claude/settings.local.json). These actions ensure the credential persists across sessions and environment reloads.
[EXTERNAL_DOWNLOADS]: Recommends the installation of the @tiny-fish/cli package from the NPM registry to provide the necessary terminal commands for the skill to function.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data fetched from external sources including arXiv, Semantic Scholar, and Google Scholar using browser automation.
Ingestion points: Data is fetched as JSON search results from multiple external web domains.
Boundary markers: No explicit boundary markers or instructions to ignore embedded malicious prompts are included in the agent's extraction goals.
Capability inventory: The skill uses the TinyFish CLI to perform network operations and shell redirection to store results in temporary files.
Sanitization: There is no evidence of sanitization or validation performed on the external content before it is synthesized into the final landscape report.
[COMMAND_EXECUTION]: Executes several shell commands to check for the presence of the required CLI tool, verify its version, and check authentication status.

academic-research-mapper