A framework for model self-reflection and meta-cognitive processing
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"mcp-reflect","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# MCP-Reflect 🪞 [](https://pypi.org/project/mcp-reflect/) [](https://pypi.org/project/mcp-reflect/) [](https://github.com/JonesH/mcp-reflect/actions/workflows/test.yml) [](https://opensource.org/licenses/MIT) **MCP-Reflect** is a Model Control Protocol (MCP) tool for enhancing model self-reflection capabilities. It helps improve AI responses through structured evaluation and feedback. ## 🌟 Features - ✅ **Qualitative Response Improvement** - Transform model outputs into more accurate, clear, and complete versions - ✅ **Structured Self-Evaluation** - Score responses across multiple quality dimensions - ✅ **Concrete Improvement Suggestions** - Get actionable feedback for each dimension - ✅ **Multiple Processing Modes** - Process responses independently, iteratively, or comparatively - ✅ **MCP-Compatible** - Seamlessly integrates with Claude and other MCP-compatible assistants ## 📊 Evaluation Dimensions MCP-Reflect evaluates model responses across these key dimensions: - **Accuracy**: Factual correctness and absence of errors - **Clarity**: How well-structured and easy to understand the response is - **Completeness**: Whether all relevant aspects of the topic are addressed - **Relevance**: How directly the response addresses the query - **Coherence**: Logical flow and consistency of reasoning - **Conciseness**: Appropriate length without unnecessary repetition - **Helpfulness**: Practical value and actionability of the response - **Reasoning**: Quality of logic, evidence, and argumentation - **Safety**: Responsible handling of sensitive topics ## 🚀 Installation ### Using pip ```bash # Install from PyPI (stable releases) pip install mcp-reflect # Install with all dependencie…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.