The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the pipedrive-pp-cli binary to interact with the Pipedrive API. It uses the Read Bash tool to execute commands for CRM management, such as pipedrive-pp-cli deals get and pipedrive-pp-cli users get.
[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of external tools from the author's own resources:
Downloads the CLI using npx -y @mvanhorn/printing-press-library install pipedrive --cli-only from npmjs.com.
Downloads and compiles the Go-based CLI and MCP server from github.com/mvanhorn/printing-press-library/library/sales-and-crm/pipedrive/cmd/pipedrive-pp-cli@latest and github.com/mvanhorn/printing-press-library/library/sales-and-crm/pipedrive/cmd/pipedrive-pp-mcp@latest.
[DATA_EXFILTRATION]: The CLI includes a --deliver webhook:<url> flag, which allows command output (potentially containing sensitive Pipedrive data like user lists or deal details) to be POSTed to an arbitrary external URL. This is a built-in feature for routing output to sinks.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes data from Pipedrive (such as notes, deal titles, and custom fields) which could contain instructions intended to influence the agent.
Ingestion points: Data retrieved from the Pipedrive API via pipedrive-pp-cli commands (e.g., notes get, deals search).
Boundary markers: The CLI uses a JSON envelope ({"meta": ..., "results": ...}) to separate metadata from the actual results.
Capability inventory: The skill can execute shell commands via Read Bash, write to local files via --deliver file:<path>, and perform network requests to Pipedrive or arbitrary webhooks via --deliver webhook:<url>.
Sanitization: There is no explicit evidence of content sanitization or instruction filtering for retrieved CRM data.

pp-pipedrive