Agent Native

vectara-hallucination-and-eval

This skill helps developers measure and reduce hallucinations in Vectara RAG pipelines using four distinct tools: HHEM for factual consistency scoring, the…

by VectaraOfficial publisherNot reviewed

Identity proven by GitHub's verified organisation badge (vectara)

Agents
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode
License
No license stated
Pin
a0b1d2bpinned Oct 9, 2026
Context cost
about 4,900 tokenswhen the skill loads
GitHub stars
0on vectara/agent-skills
Scripts
None found

What it does

“Score factual consistency with HHEM, rewrite ungrounded sentences with Vectara Hallucination Corrector, inspect query/agent traces in-console, and run offline RAG evaluation with the open-source Open RAG Eval Python toolkit. Covers when each surface is the right tool and how they compose end-to-end.”
The author's description, from SKILL.md

Generated summary from the skill's own text

This skill helps developers measure and reduce hallucinations in Vectara RAG pipelines using four distinct tools: HHEM for factual consistency scoring, the Vectara Hallucination Corrector for rewriting unsupported claims, in-console query observability for debugging, and Open RAG Eval for offline benchmark testing. It explains when to use each tool, how they compose together, and common mistakes to avoid.

When to use it

  • Score if generated text is grounded in retrieved passages with HHEM
  • Rewrite flagged generations to remove unsupported claims with VHC
  • Debug past queries or per-stage latency in the Vectara console
  • Benchmark model/reranker changes offline before deployment

When not to

  • Just building an agent or running queries without evaluation needs
  • Expecting HHEM score to indicate real-world correctness, not groundedness
  • Using VHC to invent facts missing from source documents

Install

No license stated: install from the publisher's repository, and ask the author before you reuse or share it.

These commands install from the publisher's repository and need no account.

skills CLI

npx skills add vectara/agent-skills --skill vectara-hallucination-and-eval

Installs from the repository's default branch, which may have changed since the commit we checked.

GitHub CLI 2.90 or later

gh skill install vectara/agent-skills vectara-hallucination-and-eval@a0b1d2b65a5912722493abf68243d62d5d409809

Installs the commit we checked (a0b1d2b).

Copy by hand

Download the folder at a0b1d2b and put it in .claude/skills/vectara-hallucination-and-eval for Claude Code and in .agents/skills/vectara-hallucination-and-eval for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode.

Skill folder on GitHub at a0b1d2b

Copy for your agent

A prompt for any coding agent. It installs this skill from GitHub at the pinned commit, copies the files as they are, and runs nothing unless you ask.

Prompt

Install the agent skill "vectara-hallucination-and-eval" by Vectara.

Source, pinned to commit a0b1d2b65a5912722493abf68243d62d5d409809:
https://github.com/vectara/agent-skills/tree/a0b1d2b65a5912722493abf68243d62d5d409809/skills/hallucination-and-eval

1. Download that folder at exactly this commit. Do not use the default branch.
2. Copy every file unchanged, keeping the folder structure, into .claude/skills/vectara-hallucination-and-eval/ for Claude Code and .agents/skills/vectara-hallucination-and-eval/ for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode. Create the folders if they do not exist.
3. Do not edit the files, and do not run any script in the folder unless I ask you to.
4. When you are done, list the files you copied and where you put them.

Its license is restricted or not stated. Show me the license terms from the repository before you copy anything.

Listing and checks: https://www.agentnative.dev/skills/vectara/vectara-hallucination-and-eval

SKILL.md at the pin

The license of this skill does not let us show its files here.

View SKILL.md on GitHub at a0b1d2b

Files

1 file, 20 KB in total. Each file has a sha256 in the pin's manifest; the first 12 characters are shown.

Skill folder on GitHub at a0b1d2b
All 8 skills from Vectara

More from Vectara

  • vectara-agent-auth-and-secrets

    Configure authentication for Vectara agent tools using OAuth, API keys, and service-account credentials via agent.secrets, session.metadata, and $ref…

    Claude Code, Codex, Cursor +3 · about 3.2K tokens · pinned 6 days ago

  • vectara-agent-orchestration

    This skill enables building multi-step, deterministic agent workflows using state machines with structural tool gating and platform-enforced phase transitions.

    Claude Code, Codex, Cursor +3 · about 8.6K tokens · pinned 6 days ago

  • vectara-agents

    This skill enables developers to build AI agents on the Vectara platform using retrieval-augmented generation (RAG).

    Claude Code, Codex, Cursor +3 · about 8K tokens · pinned 6 days ago

  • vectara-corpus-and-ingestion

    Manages Vectara corpus creation, document ingestion through three pathways (file upload, structured JSON, core indexing), metadata filtering with SQL-like…

    Claude Code, Codex, Cursor +3 · about 7.4K tokens · pinned 6 days ago

  • vectara-pipelines

    This skill configures Vectara pipelines, which automate bulk ingestion from external sources like S3 into Vectara agents.

    Claude Code, Codex, Cursor +3 · about 5.8K tokens · pinned 6 days ago

  • vectara-python-sdk

    The Vectara Python SDK skill helps developers build Python applications against Vectara's retrieval-augmented generation platform using typed client interfaces.

    Claude Code, Codex, Cursor +3 · about 6.5K tokens · pinned 6 days ago

All AI and agents skills

Related in AI and agents

  • trl-training

    This skill enables agents to post-train large language models using TRL (Transformers Reinforcement Learning) library, supporting supervised fine-tuning…

    by Hugging FaceOfficial publisher

    Claude Code, Codex, Cursor +4 · about 990 tokens · pinned 6 days ago

  • mcp-builder

    This skill guides agents through building Model Context Protocol (MCP) servers that connect LLMs to external services via well-designed tools.

    by SnykOfficial publisher

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 2.2K tokens · pinned 6 days ago

  • claude-settings-audit

    This skill analyzes a repository to detect its tech stack, build tools, and service integrations, then generates recommended Claude Code settings.json…

    by SentryOfficial publisher

    Claude Code · about 2.9K tokens · pinned 6 days ago

  • agents

    Build voice AI agents with natural conversation capabilities using ElevenLabs platform.

    by ElevenLabsOfficial publisher

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 6.4K tokens · pinned 6 days ago

  • agent-observability-eval-bootstrap

    Generates evaluators from production LLM traces by analyzing patterns and quality dimensions.

    by DatadogOfficial publisher

    Claude Code · about 24.5K tokens · pinned 6 days ago

  • iris-development

    This skill integrates with Redis Agent Memory, a managed service on Redis Cloud that provides persistent memory for AI agents.

    by RedisOfficial publisher

    Claude Code, Codex, Cursor +4 · about 970 tokens · pinned 6 days ago

Open roles at Vectara

No open roles at Vectara on the Agent Native jobs board right now.

Browse the jobs board

The skills listed here are third-party software, collected from public GitHub repositories as a catalog of recent skills and provided as is. Agent Native does not write or control them and is not responsible for any harm, loss or damage caused by installing or using them. Labels and checks describe what we saw at one commit and are not a guarantee. Read a skill's files before you run it. Terms of use