Agent Skills
Giskard Agent Skills give your coding agent instructions and supporting tools for working with Giskard Hub, open-source libraries, and integrations. Use them to connect an agent to the Hub, build RAG evaluations and red-team test suites, or export scan results to Collibra ↗. Each skill below includes an installation command for the skills CLI ↗ and an example prompt to get started.
To install a skill with your coding assistant, share this page and ask it to install the skill for your task.
Agent Setup v1.0.0
Section titled “Agent Setup ”Connects a local or remote agent to Giskard Hub. Your coding agent creates an authenticated HTTPS endpoint for Chat or Structured inputs, collects any streamed output into a complete response, and registers and tests the connection through the Hub SDK.
For an editable remote agent, the skill deploys the wrapper on the remote host. For a local wrapper, it opens a public HTTPS tunnel through Cloudflare or ngrok, protected by the wrapper’s API key. The skill asks for your Hub URL, Hub API key, and project name or ID.
npx skills add Giskard-AI/giskard-skills --skill hub-agent-setupIt activates on prompts like “connect my agent to Giskard Hub” or “register my chatbot in the Hub”. For example:
Connect the agent in ./agent.py to Giskard Hub in the Support project.Open Source
Section titled “Open Source”RAG Evaluator v1.1.0
Section titled “RAG Evaluator ”Turns your coding agent into an expert RAG evaluation engineer for Giskard Checks. Describe a Q&A bot grounded in documents and the skill produces a quality-focused giskard.checks evaluation suite covering groundedness, answer relevance, retrieval quality, hallucination, citation accuracy, and out-of-scope handling.
npx skills add Giskard-AI/giskard-skills --skill rag-evaluatorIt activates on prompts like “evaluate my RAG”, “test my retrieval”, “check groundedness”, or “test if my agent hallucinates”. For example:
My knowledge-base assistant in @assistant.py answers questions from documents under ./docs. Build a RAG eval suite that checks groundedness and retrieval quality.This skill is quality-focused. For adversarial or red-teaming evaluation, use Scenario Generator instead. The two are complementary and most RAG projects benefit from both.
Scenario Generator v1.1.0
Section titled “Scenario Generator ”Turns your coding agent into an expert red-teamer for Giskard Checks. Describe your agent and the failure modes you worry about, and the skill produces a complete, runnable giskard.checks test suite with adversarial scenarios and layered checks.
npx skills add Giskard-AI/giskard-skills --skill scenario-generatorIt activates on prompts like “create test scenarios for my bot”, “red-team my RAG system”, or “generate checks for prompt injection”. For example:
My customer support bot available on @support_bot.py must never leak customer PII or discuss competitors. Generate a red-team suite.Integrations
Section titled “Integrations”Giskard to Collibra v1.0.0
Section titled “Giskard to Collibra ”Exports Giskard Hub scan results into Collibra AI Governance ↗. The skill finds the scan in Giskard Hub, creates or updates the AI Agent asset hierarchy in Collibra (re-runs are idempotent), and pushes per-probe pass/fail metrics to the Quality tab. It requires a Collibra instance with AI Governance.
npx skills add Giskard-AI/giskard-skills --skill giskard-to-collibraIt activates on prompts like “export my scan to Collibra”, “push results to Collibra”, or “send the latest scan of this project to Collibra”. For example:
Retrieve the latest scan of the Zephyr project and send it to Collibra.Install as a Claude Code plugin
Section titled “Install as a Claude Code plugin”The repository also works as a Claude Code plugin ↗. Clone it and start Claude Code with the --plugin-dir flag:
git clone https://github.com/Giskard-AI/giskard-skillsclaude --plugin-dir ./giskard-skillsOnce Claude Code is running, use the /giskard-skills command to list the available skills.