DEV Community

Model Context Protocol

MCP is an open protocol that standardizes how applications provide context to LLMs.

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Tool count is a vanity metric. Annotation coverage is what makes an AI agent safe.
favcrm profile
FavCRM

Tool count is a vanity metric. Annotation coverage is what makes an AI agent safe.

#ai #mcp #security #webdev
1
Comments
3 min read
Your MCP Client Runs the Command Before It Trusts the Server
jfisher4002 profile

Your MCP Client Runs the Command Before It Trusts the Server

#security #ai #testing #mcp
1
Comments 3
7 min read
What is an 'agentic CRM'? (and why MCP made it real)
favcrm profile
FavCRM

What is an 'agentic CRM'? (and why MCP made it real)

#ai #mcp #crm #saas
1
Comments
2 min read
Obot Platform v0.22.0: Centrally Managed Skills, Fleet Scanning & Enterprise Controls
obot_ai profile

Obot Platform v0.22.0: Centrally Managed Skills, Fleet Scanning & Enterprise Controls

#mcp #opensource #devtools #ai
1
Comments
1 min read
We collapsed 76 MCP tools into 12 — and cut per-session context cost by 60%
dorikuio profile
Doriku

We collapsed 76 MCP tools into 12 — and cut per-session context cost by 60%

#mcp #ai #devtools #productivity
1
Comments 8
3 min read
New Zealand Companies Office — what the register actually returns
openregistry profile

New Zealand Companies Office — what the register actually returns

#programming #mcp #kyb #legaltech
1
Comments
3 min read
"It's not a bug, it's spec": a zero-click RCE in AI coding agents that three vendors won''t patch
ju571nk profile

"It's not a bug, it's spec": a zero-click RCE in AI coding agents that three vendors won''t patch

#ai #security #mcp #devsecops
1
Comments
6 min read
I shipped 7 production MCP server Actors in two weeks — here's what the docs don't tell you
unbearablelabs profile

I shipped 7 production MCP server Actors in two weeks — here's what the docs don't tell you

#mcp #apify #python #devops
1
Comments
8 min read
My AI memory benchmark said 98.3%. The number was true — and worthless.
daniel_nevoigt_ca2fdc23d5 profile

My AI memory benchmark said 98.3%. The number was true — and worthless.

#ai #mcp #opensource #benchmarking
Comments 14
4 min read
"Put a security gateway in front of any MCP server in 5 minutes"
c6zks4gssndroid profile

"Put a security gateway in front of any MCP server in 5 minutes"

#ai #mcp #python #security
2
Comments
5 min read
I got tired of re-explaining my codebase to every coding agent — so I made critical memory live in the repo next to code
xchucx profile

I got tired of re-explaining my codebase to every coding agent — so I made critical memory live in the repo next to code

#ai #mcp #opensource #go
Comments
8 min read
Talon: an open-source agentic AI harness that lives across Telegram, Discord, Teams & your Terminal
talon_agent profile

Talon: an open-source agentic AI harness that lives across Telegram, Discord, Teams & your Terminal

#ai #opensource #mcp #typescript
2
Comments
2 min read
I gave my AI a "publish this" tool over MCP. Here is what that tool actually has to do.
thryvate profile
Alex

I gave my AI a "publish this" tool over MCP. Here is what that tool actually has to do.

#ai #webdev #devtools #mcp
2
Comments 3
4 min read
From dynamic to adaptive: rewriting an agent's reasoning operation to its exact task at runtime
frank_brsrk profile

From dynamic to adaptive: rewriting an agent's reasoning operation to its exact task at runtime

#ai #llm #mcp #agents
Comments
2 min read
Anthropic Published a 31.5% Hijack Rate. Most Vendors Won't Even Show You a Number.
agentshield profile

Anthropic Published a 31.5% Hijack Rate. Most Vendors Won't Even Show You a Number.

#aisecurity #llm #promptinjection #mcp
Comments
5 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.