pp-discogs
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires downloading a command-line tool and an NPM package from the vendor's official repositories on GitHub (
github.com/mvanhorn/printing-press-library) and the NPM registry (@mvanhorn/printing-press-library). - [REMOTE_CODE_EXECUTION]: Installation instructions include executing remote code via
go installfor the Discogs CLI andnpxfor the Printing Press installer package. - [COMMAND_EXECUTION]: The skill operates by executing shell commands through the
discogs-pp-clianddiscogs-pp-mcpbinaries. - [DATA_EXFILTRATION]: The CLI supports a
--deliver webhook:<url>flag, allowing command output (which may contain collection or marketplace data) to be transmitted to a remote HTTP endpoint. - [CREDENTIALS_UNSAFE]: The skill manages a personal access token (
DISCOGS_TOKEN) and mentions storing sensitive credentials in a localcredentials.tomlfile or environment variables. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes data from the Discogs API and a local SQLite database to perform collection management and marketplace decisions.
- Ingestion points: Data enters the agent context from Discogs API responses and a local SQLite database (
data.db). - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the provided templates for processing API data.
- Capability inventory: The CLI can perform network operations (API calls, webhooks), write to the local filesystem (database and configuration), and execute shell commands.
- Sanitization: Documentation mentions a PII stripping mechanism for its internal logs, but no specific sanitization is detailed for data processed during routine command execution.
Audit Metadata