pp-espn

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the espn-pp-cli tool via npx from @mvanhorn/printing-press-library or go install from github.com/mvanhorn/printing-press-library. These resources are authored by the skill's developer.
  • [COMMAND_EXECUTION]: Instruction templates like espn-pp-cli recall "<user's question>" incorporate user input directly into shell commands, creating a command injection risk if the agent fails to escape input. The skill also recommends backgrounding commands with the shell & operator.
  • [DATA_EXFILTRATION]: The tool supports routing output to external webhooks via the --deliver flag and sending diagnostics via a feedback command to a configurable ESPN_FEEDBACK_ENDPOINT.
  • [DYNAMIC_EXECUTION]: A sql command allows arbitrary SQL queries against a local SQLite database, enabling dynamic code execution within the skill's operational environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from the ESPN API and local database, which is interpolated into the agent's context. 1. Ingestion points: ESPN API responses and local SQLite database. 2. Boundary markers: None specified. 3. Capability inventory: Bash tool used to run CLI with network, file write, and SQL capabilities. 4. Sanitization: Not specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:45 PM