The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the deepsource CLI to perform various actions such as authentication (deepsource auth login) and retrieving analysis data (deepsource issues, deepsource report-card, deepsource vulnerabilities). These are standard operations for the intended use case.
[EXTERNAL_DOWNLOADS]: The skill references the official installation documentation at https://deepsource.com/docs/developers/cli/installation. This is a link to the vendor's own website for utility setup.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from DeepSource analysis reports which could theoretically contain malicious instructions if the source repository or DeepSource account is compromised.
Ingestion points: Data enters the context through deepsource issues, deepsource report-card, deepsource vulnerabilities, and deepsource runs commands in SKILL.md.
Boundary markers: None present; the JSON output is parsed directly into the agent's context.
Capability inventory: The skill primarily executes the deepsource CLI tool as defined in SKILL.md.
Sanitization: No explicit sanitization or filtering of the CLI's JSON output is defined in the instructions.

deepsource