Skip to main content
Glama
Fredo QA Skills MCP Server

Fredo QA Skills MCP Server

by illicomedia

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
FREDO_QA_API_URLNoCustom API gateway URL.
FREDO_QA_LICENSE_KEYNoInline signed license key or license content for the active session.
FREDO_QA_LICENSE_FILENoPath to a signed Ed25519 license file (license.json) used to unlock Pro and Enterprise skills.

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
list_skillsA

List all Fredo QA Skills available under the current license tier (Free, Pro, Enterprise), with rule summaries and categories.

get_skill_rulesA

Return the SKILL.md of a Fredo QA skill installed in the working directory (skills// or .agents/skills//). This server ships no skill files: if the skill is not installed, the call fails with installation instructions; do not invent rules for it.

install_skillA

Install a Fredo QA skill into .agents/skills/<skill_id>/ with SKILL.md and LICENSE, copying SKILL.md from a skill source already present in the project (skills// or .agents/skills//); fails if none exists. Requires user agreement: set confirm: true to write files to disk; when false (default), returns a safe preview without modifying the filesystem.

sync_workspace_skillsA

Auto-detects project stack and provisions matching QA skills and licenses into .agents/skills/. Requires user agreement: set confirm: true to write files; when false (default), returns a dry-run preview.

prune_skillsA

Safely prunes unneeded QA skills from .agents/skills/ by archiving them to .agents/.archive/ based on detected project tech stacks. Non-destructive: files are moved rather than permanently deleted, keeping them fully recoverable while removing noise from the AI agent context. Requires user agreement: set confirm: true to archive skills; when false (default), returns a safe preview listing candidate skills without moving files.

set_license_keyB

Set the signed Fredo QA Skills license file for the active session to unlock Pro or Enterprise skills.

verify_licenseA

Inspect the current Fredo QA license: signature validity, tier, licensee, last covered release date (updates_until), and fallback status.

quick_checkA

Sub-second preflight check for AI agents to discover repository stack, governance policies, and baseline scaffolding before modifying code. Verifies presence and configuration of agent instructions (AGENTS.md), licensing, secret hygiene (.gitignore), dependency lockfile, and test runner declaration. Requires user agreement: set confirm: true to perform directory traversal and inspect project files; when false (default), returns a preview describing the audit scope and target directory without scanning the filesystem.

get_versionA

Retrieve the active package version, protocol specification, and runtime metadata of the Fredo QA MCP Server.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

TDQS

A3.8/5.0

Scored across 9 tools

Disambiguation4/5

Most tools target a distinct action: list_skills (enumerate), get_skill_rules (read one skill's SKILL.md), install_skill (single install), sync_workspace_skills (bulk auto-provision), prune_skills (archive). The main friction is install_skill vs sync_workspace_skills, and quick_check vs sync_workspace_skills, both of which detect project stack; descriptions differentiate them, but an agent could still hesitate over which provisioning path to pick.

Naming Consistency4/5

Eight of nine tools follow a clean verb_noun snake_case pattern (list_skills, get_skill_rules, install_skill, set_license_key, verify_license, get_version). The one deviation is quick_check, which is adjective_noun rather than verb-led, a minor inconsistency in an otherwise predictable scheme.

Tool Count5/5

Nine tools is well within the ideal 3-15 range and each earns its place across the three sub-domains (skill inventory/provisioning, licensing, preflight audit, version metadata). No filler or redundant tools inflate the surface.

Completeness4/5

The skill lifecycle is largely covered: list, read rules, install, bulk sync, and prune (non-destructive removal), plus license set/verify and version info. Minor gaps remain — no explicit uninstall/disable, no skill update or authoring path, and no way to filter skills by tier/category in list_skills — but agents can work around these via prune and list_skills.

Maintenance

ActivityActive
ResponsivenessNo recent community issues