skills.rest
emirhankudun-ux
emirhankudun-uxCOMMUNITY
Explore all skills by this author →
GitHub

seis-mcp-runtime

Audits MCP endpoints, tool boundaries, and runtime evidence for secure AI operations.

1|Updated Jun 5, 2026
Search Tags:#security#audit#ai agents#mcp#endpoint verification#runtime validation#tool boundaries
One-click install
npx skills add https://github.com/emirhankudun-ux/SEIS --skill seis-mcp-runtime-emirhankudun-ux
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: seis-mcp-runtime
Source: https://github.com/emirhankudun-ux/SEIS/tree/main/plugins/seis/skills/seis-mcp-runtime
Command: npx skills add https://github.com/emirhankudun-ux/SEIS --skill seis-mcp-runtime-emirhankudun-ux

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams publishing MCP capability claims often rely on stale manifests or unverified assumptions, leading to inaccurate runtime readiness statements and security boundary drift. This Skill validates MCP tool inventories against declared lane ownership and security boundaries before any claim goes live. ## Core Features & Use Cases - MCP Endpoint Audits: Reads local MCP manifests and runtime check scripts to verify tool availability and resource exposure. - Tool-to-Lane Traceability: Compares declared lane ownership against actual tool boundaries to detect drift. - Evidence Mapping: Attaches runtime verification evidence and remediation actions to every capability claim. - Use Case: Before publishing a Command Center dashboard claiming live MCP tool access, run this Skill to confirm each tool is verified, scoped, and free of leaked secrets. ## Quick Start Use the seis-mcp-runtime skill to audit my MCP tool inventory and verify runtime readiness before publishing capability claims.

Frequently Asked Questions about seis-mcp-runtime

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit MCP tool boundaries before publishing runtime claims?▼

Read local MCP manifests and runtime check scripts, then compare tool availability against declared lane ownership and security boundaries. Attach verification evidence and remediation actions to any claim surface before publishing.

What validation commands check MCP runtime readiness?▼

Run npm run check:plugin-capability-lanes, check:seis-agent-plugin-integration, check:seis-ai-agent, and check:seis-specialist-plugins. These scripts validate tool inventories, lane traceability, and integration boundaries.

Can I claim live external MCP access from static manifests?▼

No. Static manifest copies alone are insufficient evidence for readiness claims. Mark capabilities as planned until active verification confirms the tool, resource, or endpoint is live and scoped correctly.

How does the skill handle API keys and endpoint secrets?▼

It enforces a zero-key principle for default demo mode and prohibits exposing private tool keys, API tokens, or endpoint secrets in prompt surfaces. Evidence stays attached to the tool that produced it without leaking credentials.

What happens when an MCP tool or endpoint is stale?▼

The workflow identifies drift between declared and actual capability, then attaches clear remediation next actions to the claim surface. Stale tools are flagged rather than silently published as ready.

Related Skills

Explore More Software Engineering Skills→
julianjabjulianjab

new-mcp

Scaffolds and audits MCP server plugin skeletons with manifests and template generation.

Community
Advanced
manatagomanatago

kiro-verify-completion

Verifies task, fix, test, and feature claims using fresh command outputs and exit codes.

Community
Advanced
latestaiagentslatestaiagents
5

secrets-detection

Detect leaked API keys and credentials in code repositories.

Official
Intermediate
intersystems-communityintersystems-community
27

objectscript-tdd

Automate the compile-test-fix loop for ObjectScript classes.

Official
Intermediate