MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
tibet-toolsgithub.com/jaspertvdm/tibet-toolsupdated this year
MCP tools for Claude Code — browse, install, and manage 20+ AI tools with TIBET provenance
pypi MIT -
librarian-mcpgithub.com/liana-banyan/librarian-mcpupdated this year
Pre-curated canonical memory + prose/code provenance for any AI coding client. MCP-native.
pypi AGPL-3.0-or-later -
Governed, AI-safe access to customer data — an MCP server enforcing consent, PII removal, provenance, and audit on every agent read. Spec-first with Claude Code, 99 tests.
MIT -
inaligngithub.com/Intellirim/inalignupdated this year
AI Agent Governance Platform — Cryptographic provenance chains, behavioral analysis, and tamper-proof audit trails for AI coding agents. MCP native.
pypi MIT ★ 2 -
Relational codebase compiler and spec-driven containment engine for AI coding agents. Compiles polyglot repos (Rust, Python, TS, Go, C/C++, C#, Lean 4) into an AST-proven Datalog & Louvain index—enforcing virtual-AST pre-flight specs, zero-config modularity (ΔQ), and scope-creep gates via MCP & CI.
MIT ★ 32 -
Persistent memory for AI agents and coding assistants: exact versions, provenance, withdrawal with cascade, MCP server and Hermes Agent provider
pypi MIT -
Turn documents, wikis, emails, and Confluence/SharePoint/OneDrive/Google Drive sources into governed, provenance-bearing knowledge AI agents can trust — reviewed publication, conflict retention, citation-exact retrieval. Native in Claude Code, Codex, Kiro, and any MCP client.
MIT -
Personal, local-first conversation memory and evidence retrieval for AI coding assistants, with provenance, evaluation, and recovery.
MIT -
seif-mcpgithub.com/and2carvalho/seifupdated this year
MCP server for the SEIF Protocol — provenance signing, transparency log, and data classification tools for AI agents (Cursor, Claude Code, Claude Desktop). Thin Node wrapper around the canonical Python implementation.
npm MIT -
Local-first project continuity for AI coding agents: durable memory, provenance and structured handoffs across Claude Code, Codex and new sessions.
Apache-2.0 -
codebase-intelgithub.com/mutharasu/codebase-intelupdated this year
Your AI agent writes code — but does it know WHY your code exists? Decision provenance, quality contracts, and AI anti-pattern detection for coding agents.
pypi MIT -
astramem-localgithub.com/astragenie/astramem-localupdated this year
Local-first, encrypted memory daemon for AI coding agents — hybrid retrieval, provenance receipts, MCP tools, SQLCipher at rest
npm MIT -
Your AI agent writes code — but does it know WHY your code exists? Decision provenance, quality contracts, and AI anti-pattern detection for coding agents. 19 languages. MCP compatible.
MIT ★ 5 -
ai-productivity-dashboardgithub.com/riko5652/ai-productivity-dashboardupdated this year
AI memory engine — learns from your coding sessions, recommends tools, injects proven solutions via MCP. Tracks Claude Code, Cursor, Aider, Windsurf, Copilot & Gemini. No API keys required.
npm AGPL-3.0-or-later -
The full-state-verification layer for AI coding agents: one native MCP server (Rust host, C engine) that turns a repository into measured, grounded, guarded, predictive, self-improving intelligence with provenance for every claim.
NOASSERTION ★ 5 -
Cross-agent memory for AI coding agents. Verbatim transcripts + distilled memory with provenance, hub/edge replication, MCP tools. Rust · SQLite · local-first · MIT.
MIT -
Agent Skills + MCP servers for AI coding agents (Claude Code, GitHub Copilot, Codex, Cursor, Gemini CLI): bug analysis, regression tracing with git bisect, proven fixes, security audit, APK reverse engineering, ticket reading.
MIT ★ 1 -
The ultimate tree-sitter-backed patching engine for AI agents. Safely apply LLM-generated code changes across 16+ languages with AST-aware validation, fuzzy SEARCH/REPLACE, cryptographic SCITT provenance, and native MCP integration. 🤖🌳⚡
NOASSERTION ★ 1 -
Structured, provenance-aware context for AI agents, built from databases, source code and production systems
MIT -
The black box for AI-written code. Passive decision graph linking every commit to the agent prompts behind it: drift detection, DLP audit of what agents read, evidence-driven risk, MCP memory for coding agents, signed provenance attestations. One command, offline by default.
Apache-2.0 ★ 289 -
MCP server that keeps account of what AI coding agents leave behind on a Mac: processes, ports, temp files, Docker objects, simulators, emulators, caches. Removes only what is proven theirs. Native AOT, stdio.
nuget MIT -
Local-first shared memory for AI coding agents — one ambient memory across Claude Code and Codex: session briefings, pitfall warnings before repeated mistakes, auto-learned lessons with per-agent provenance, and post-compaction recovery. MCP server + life
npm Elastic-2.0 -
Local-first shared memory for AI coding agents — one ambient memory across Claude Code and Codex: session briefings, pitfall warnings before repeated mistakes, auto-learned lessons with per-agent provenance, and post-compaction recovery. MCP server + life
npm Elastic-2.0 -
chainsightgithub.com/YashwinReddy29/chainsightupdated this year
First CycloneDX 1.6 SBOM tool for AI generated code. Intercepts IBM Bob IDE sessions through 4 MCP tools, catches CVEs past the model training cutoff, and classifies code provenance with LLaMA 3.3 70B on watsonx.ai.
-
varvegithub.com/varve-sh/varveupdated this year
Decision memory for AI coding agents. A local-first Go binary and MCP server: records decisions with evidence, scope, provenance and expiry, packs the relevant ones into each agent session within a token budget, and links commits back to the decisions that were in scope when they landed.
go Apache-2.0 ★ 9 -
OriginTrail Decentralized Knowledge Graph (DKG) — Shared memory for AI agents. Give Hermes, OpenClaw, Claude Code, Codex, and Cursor a local knowledge-graph memory that several agents can share, with provenance, and verify on-chain when it matters. Open source, Apache-2.0.
go Apache-2.0 ★ 136 -
Self-hosted memory + evidence layer for AI agents (Claude Code, Codex, Hermes, ...) — embeddable Go library, MCP / HTTP / CLI, evidence-backed claims, bitemporal recall, axi-go execution kernel with JSONL audit + token budgets, cosign-signed releases with SLSA L3 provenance. No vendor cloud, no per-call billing.
MIT ★ 5 -
EU AI Act and HLEG Trustworthy AI as a provenance-carrying knowledge graph, served to coding agents through MCP tools. Engineering and documentation support, not legal advice.
AGPL-3.0 ★ 1 -
Help AI coding agents avoid repeated mistakes, reuse proven fixes, and share practical lessons across sessions and tools. Local memory with automatic recall in supported hosts and recommendations informed by verified results.
npm MIT -
Security layer for AI agents: real-time interception of pip/npm/cargo/gem/go installs via Hermes & OpenClaw plugins, Claude Code/Codex hooks, MCP, shell guard, PATH shim, execve, and index proxy. Checks CVEs (OSV/NVD/GitHub), typosquats, malware, licenses, provenance, trust, lockfile hashes. Offline mirror, hash-chained audit, SBOM/SARIF.
MIT ★ 4 -
Read-only, local-first dashboard showing what your AI agents (Claude Code, Gemini CLI, MCP servers) are actually allowed to do: permission rules with file provenance, MCP server inventory, sandbox & hook posture, risk findings, OWASP Agentic Top 10 threat scenarios, and config drift. No enforcement, no telemetry (for now).
Apache-2.0 -
Fledly for AI agents: iOS App Store market data (young apps that earn, proven app ideas, top charts, keyword research, rank tracking) via the hosted Fledly MCP server. Claude Code, Cursor, Hermes, OpenCode and npx skills.
MIT -
AI-driven DFIR framework: an MCP server that turns Claude Code into an autonomous digital-forensics analyst on SANS SIFT - correlates disk + memory evidence, enforces a deterministic evidence-provenance gate, and produces auditable, hash-chained investigation reports.
MIT ★ 5 -
neruva-mcpupdated this year
Neruva MCP -- proprietary memory + grounding substrate for AI assistants (Claude Desktop / Claude Code / any MCP client). Typed Records store, document knowledge graph with exact multi-hop traversal (ownership/relationship tracing), semantic + structural recall, deterministic corrections, provenance/citations, GDPR/Law-25 atomic forget, snapshot/replay audit. Thin client: all algorithms run server-side. For Neruva integration clients only -- NOT open source. Bring your own LLM.
pypi Proprietary -- (c) Clouthier Simulation Labs. For Neruva integration clients onl -
A judge for code written by AI: plain Python, rules against hallucinations.
pypi MIT -
MCP server for the three.ws append-only, signed agent action log: record, list and verify actions.
Apache-2.0 ★ 9 -
Agent-governed job-search CRM: Claude agents verify roles at each employer’s ATS, score fit and recommendations under scoped credentials, with an append-only audit log. Rails 8 + Python/Playwright + MCP.
MIT ★ 1 -
The model proposes, deterministic code disposes, a human signs. Content-addressed envelopes, hash-chained audit, signed approval — for ledgers, Lean proofs, manuscript claims, reproductions, benchmarks, peer review. Conformance-tested; byte-compatible with @kernel.chat/kbot-finance.
MIT -
knowbasegithub.com/v0ropaev/knowbaseupdated this year
Git-versioned, provenance-grounded knowledge layer extracted from a codebase, served to humans and AI agents via MCP
AGPL-3.0 ★ 9 -
TRACE protocol for auditing decision-provenance in AI-assisted scientific workflows
Apache-2.0 ★ 26 -
Memory server for MCP clients (TypeScript, SQLite/FTS5): quarantines injected instructions and records provenance; on npm.
npm MIT -
@getnema/mcpgithub.com/albertogrande/nemaupdated this year
Nema MCP server: read tools + write tools (draft/update/propose/check/request_review)
npm Apache-2.0 -
MCP server exposing Cortex engineering state and handoffs to coding agents
npm MIT -
Lossless, compressed, verifiable memory for AI agents.
pypi LicenseRef-Proprietary -
A claim–evidence ledger for AI agents (CLI · MCP · hooks): re-checks evidence and flags what breaks
Apache-2.0 ★ 3 -
altretta-skillgithub.com/ApiliumCode/altretta-skillupdated this year
Give your AI a second brain. Altretta connects your local, encrypted knowledge vault to 70+ AI agents with query-first, cited retrieval and signed provenance.
MIT -
Turn Claude Code into an interactive scientific workspace: workflow, file system, and knowledge graph.
pypi MIT ★ 14 -
Verify agent-authored change claims at exact Git commits with an evidence-first PR gate and local MCP.
npm MIT -
cogmemgithub.com/writerslogic/cogmemupdated this year
A self-improving, verifiable memory layer for AI coding agents.
pypi Apache-2.0 -
MCP server for Proof: an audit trail for AI-written code. Records what the agent changed, what a person actually reviewed, and who signed off. Works inside Claude Code, Claude Desktop and Cursor.
npm AGPL-3.0 -
Fail-closed evidentiary eligibility engine for AI-agent norms, with a self-linting claim map.
Apache-2.0 ★ 2 -
COGO — la memoria con semáforo de confianza para construir software con IA. Suite Escriba.
MIT ★ 1 -
Paused research project: an MCP research workspace for AI agents with persistent evidence, counterevidence and gap tracking. Evaluation results in README.
pypi MIT ★ 2 -
Self-hosted memory for AI agents via MCP. Capture source events, retrieve context across sessions, and explain evidence-backed claims.
pypi MIT ★ 8 -
Persistent decision memory for AI agents: rationale, rejected approaches, and entity history. Connect any compatible agent through local MCP or the CLI.
pypi MIT ★ 26 -
Evidence-chain whiteboard + MCP server for AI research agents: every claim traces back to the run and code that produced it. 讓 AI agent 做研究時,每個結論都能追到產生它的程式碼。Claude Code · Codex · Cursor · Gemini CLI
npm AGPL-3.0 ★ 501 -
Unofficial evidence-first CLI + read-only MCP server for X's open-source algorithm (xai-org/x-algorithm): commit-pinned SHA-256 spans, code index, findings that go STALE.
Apache-2.0 -
Local-first memory for AI agents. Keep decisions and evidence across sessions. Recall with sources through MCP.
cargo AGPL-3.0 ★ 9 -
Local-first toolkit: WhatsApp order messages become source-linked drafts a person checks and exports. MCP server for coding agents. Early prototype, synthetic data only.
MIT -
Structured, portable, auditable memory for AI agents — plain Markdown (.mem.md) you can read, diff and audit. Spec + CLI + MCP server for Claude Code, Cursor, and any MCP client.
npm Apache-2.0