More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Repair RAG Retrieval, Grounding & Citation Failures

    1

    Repair RAG Retrieval, Grounding & Citation Failures is a bounded ToolForge Labs workflow.

    $1

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Repair RAG Retrieval, Grounding & Citation Failures

    Repair RAG Retrieval, Grounding & Citation Failures

    Example session with this skill installed

    Our RAG assistant sometimes cites the wrong document, retrieves stale chunks after a source update, and misses a clearly relevant passage when metadata filters are enabled. Repair retrieval and citation binding without relaxing tenant access controls.

    • Read your context and instructions
    • Compiled the repair rag retrieval

    The likely issue is a contract mismatch around RAG retrieval, chunking, filters, embeddings/index versions, ranking, context assembly, grounding, citations, stale content, access control, and evaluation. I would first define the invariant: A generated claim may cite a source only when the exact context fragment supplied to the model can be traced to an authorized current source and supports that claim; retrieval misses and unsupported citations must remain observable. Then I would trace the smallest owning boundary, apply or propose a bounded repair/review step, and add deterministic normal, boundary, failure and replay tests. The workflow preserves source access controls, privacy, document ownership, citation/provenance requirements, and product answer policy. If repository/workspace files are available, the findings can be mapped to exact paths and focused local validation; with partial evidence, the result stays an explicit diagnosis/evidence map with assumptions and no invented execution claims.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Repair RAG Retrieval, Grounding & Citation Failures is a bounded ToolForge Labs workflow. Diagnose, review, and repair RAG retrieval, chunking, filters, embeddings/index versions, ranking, context assembly, grounding, citations, stale content, access control, and evaluation. Works from repositories/workspaces or partial supplied evidence. Produces bounded fixes or evidence gates with deterministic validation while preserving source access controls, privacy, document ownership, citation/provenance requirements, and product answer policy and avoiding invented execution claims. It works from a full repository/workspace or partial evidence such as snippets, logs, configuration, traces, documents, tickets, and architecture notes. The skill separates verified facts from assumptions, avoids invented execution claims, and produces concrete repair/review steps with focused deterministic validation or traceable evidence.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.

    Requires either repository access or specific logs/traces to be effective. Cannot verify live production system state without provided evidence.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    Use a clean branch or controlled review workspace. Supply repository/workspace files, logs, traces, configuration, documents, or evidence needed for the task. Review patches/findings before merge or operational action; keep production credentials and destructive production actions out of the workflow.

    Listed1 month ago
    Updated9 days ago

    What's inside

    Frequently Asked Questions

    This skill provides an end-to-end diagnostic and repair workflow for common RAG issues like "hallucinations," broken citations, poor retrieval ranking, and improper document chunking. It helps teams move from unreliable, ungrounded AI responses to traceable, deterministic outputs.

    Yes, the workflow is designed to process either full codebase repositories or partial evidence like logs, vector index configurations, and trace snippets. This allows you to diagnose systemic failures even if you cannot provide a live connection to your production database.

    The skill maintains strict adherence to source access controls and document ownership, ensuring that repair recommendations do not bypass your existing privacy filters or security protocols. It focuses on deterministic validation to ensure data remains within your defined boundaries.

    This is a ToolForge Labs workflow compatible with any AI agent or platform that supports external skill integration via API. It is designed to be model-agnostic, working across various vector databases, embedding models, and LLM orchestration frameworks.

    Unlike general debugging tools, this skill generates specific "evidence gates" and deterministic validation steps. This means every fix is backed by traceable provenance, preventing the AI from making "invented execution claims" about what was repaired.

    New in LLM Ops