firm-spec-compliance-pack_skill

This skill helps ensure MCP 2025-11-25 specification compliance across elicitation, tasks, prompts, audio, JSON schema, SSE, and icon metadata audits.

Python

2.5k

GitHub Stars

2

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill openclaw/skills --skill firm-spec-compliance-pack

_meta.json313 B
SKILL.md2.4 KB

Overview

This skill validates an agent's implementation against the MCP 2025-11-25 specification. It bundles seven focused audits that check elicitation, task handling, resources/prompts, audio content, JSON Schema 2020-12 dialect, SSE transport resumption, and icon metadata. Use it to catch protocol non‑conformance before deployment.

How this skill works

Each audit tool runs automated checks against your agent configuration and runtime artifacts. Tools parse configuration files, inspect declared capabilities and handlers, validate message formats (including binary audio blobs), verify JSON Schema dialect URIs and vocabularies, and simulate SSE resumption scenarios. Results are returned with severity levels and actionable findings to guide fixes.

When to use it

Before releasing a new agent or skill version that targets MCP 2025-11-25
During CI pipelines to enforce protocol compliance gates
When adding or changing elicitation flows or long-running task handlers
After integrating audio content or updating JSON Schemas
When implementing SSE streaming or icon metadata handling

Best practices

Run the full pack in CI to prevent regressions and enforce compliance
Provide a complete agent configuration JSON to get accurate checks across tools
Prioritize HIGH severity findings (elicitation, tasks, JSON Schema, SSE) for production deployments
Keep mcp-openclaw-extensions dependency >= 3.0.0 and target protocol version 2025-11-25
Review human-flagged items; the pack flags issues but requires human validation before changes are applied

Example use cases

Validate a new elicitation flow to ensure server-initiated prompts conform to SEP-032
Audit task implementation to confirm durable request lifecycle and status reporting per SEP-044
Verify JSON Schemas use the 2020-12 dialect and required vocabularies
Test SSE event streaming and Last-Event-ID resumption behavior in staging before go‑live
Check icon URIs and media types to meet accessibility and metadata rules

FAQ

No. The pack reports findings and recommendations. Human review is required before applying changes.

What input does each audit expect?

Each tool accepts a path to your agent configuration (config.json) and inspects declared handlers, schemas, and metadata. Provide complete runtime artifacts for best results.