MCP server that translates a lockfile diff (npm, PyPI) into a
human-readable upgrade plan — semver class, breaking changes from
GitHub release notes, CVEs fixed in range, migration guide links,
and a per-package recommendation. Install via npx.
Adds toolstem/toolstem-mcp-server — agent-ready financial intelligence MCP with derived signals (UNDERVALUED/STRONG/ACCELERATING) and pre-computed CAGRs/margin trends/DCF upside. Published to npm, the Official MCP Registry, and the Apify Store.
Non-custodial on-chain escrow + AI arbitration for A2A USDC payments on Base.
Wraps EscrowV1 contract on Base Sepolia with 60 tools for buyer/seller/arbiter flows.
Per github-actions[bot] comment on this PR: add the Glama MCP
server score badge. Placement matches the convention used by
other Knowledge & Memory entries on this list — right after
the GitHub repo link, before the emoji flags + description.
Underlying server:
- Dockerfile added upstream at Battam1111/Myco@dec7f6c so
Glama's introspection check can start + verify the server.
- GitHub topic `mcp` present (one of 20 topics).
- Listed in the official MCP Registry as
`io.github.Battam1111/myco` v0.5.12.
Glama may still be crawling; the badge will display the
placeholder score until the crawler indexes Battam1111/Myco.
Tap compiles AI browser automation into deterministic .tap.json plans
(25-op closed union, zero runtime LLM), runs on the user's logged-in
Chrome, and detects drift via semantic fingerprint diff.
Distinct from neighbors:
- opentabs-dev/opentabs: authenticated browser access (no compile step)
- browsermcp/mcp: remote-control only (no plan/lifecycle)
- playwright-mcp: structured accessibility, no plan compilation
- Libretto / BrowserBook: compile but no drift detection / lifecycle
Glama score pending evaluation (Dockerfile fix just pushed).
Resubmission of closed PR #4148 after Glama scoring completes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Battam1111/Myco — agent-first cognitive substrate. 18 manifest-driven verbs + 25 lint dimensions enforcing contract invariants mechanically. Cross-session / cross-project memory via a self-validating filesystem graph (AST + markdown-link derived, not embedding-based). Provider-agnostic: MP1/MP2 dims forbid LLM-SDK imports in kernel and plugins.
Python, local service, cross-platform. MIT-licensed. Works with Claude Code, Cursor, Windsurf, Zed, VS Code, and any MCP client.
Creates a new Conversational AI category (placed alphabetically between
Communication and Customer Data Platforms) and populates it with the
Perspective AI MCP server.
Perspective AI's AI Concierge replaces static forms with adaptive AI
conversations for lead qualification, customer research, onboarding
feedback, and advocacy. Published on the official MCP Registry at
ai.getperspective/mcp.
* Add Glama score badge as requested by the bot (resolves
`missing-glama` label once the server is indexed by Glama).
* Drop mention of "financial news" and loose "public endpoints
need no key" phrasing. v0.2.0 removed the news tool and now
gates historical rates behind ALLRATES_API_KEY — only the
simple-rate endpoint is fully public.
26-tool Substack MCP server with no-API-key reply drafting via
propose_reply / confirm_reply pattern. Host LLM (Cursor / Claude
Desktop / Claude Code) does the writing.
uvx-installable one-liner with safety rails (dedup, audit, dry-run
default, rate limiter).
Frank Fiegel
Kaustubh Ghodke
omniologynow-rgb
Hirokazu Yoshinaga
SportScore
toolstem
Depscope
Mr Shannon
UPinar
Calum Roberts
Catrin Donnelly
JohnnyLiangtyxuan
Battam1111
LeonTing1010
Battam1111
Paul Burgess
Reynier
MastadoonPrime
Alex
José Manuel Requena Plens
Igor Peregudov
8randonpickart5
vinaypuppal
cahthuranag
06ketan
Ketan Chavan
Alex Chernysh
