Automation
Plugin
claude-prompts
A Model Context Protocol (MCP) server for prompt workflows. Written once, always followed. Craft reusable prompt templates with quality gates and reasoning guidance. Orchestrate multi-step workflow chains with a composable operator syntax.
Plugin Signal75/100Solid
maintained100
capable43
documented100
adopted48
Stats
185
Stars
33
Forks
Active
Maintenance
TypeScript
Language
MIT
License
2h ago
Last commit
1y ago
Created
Repo: minipuft/claude-prompts-mcp
What's inside
Hooks
5Badges
4
README
1Claude Prompts
A Model Context Protocol (MCP) server for prompt workflows. Written once, always followed.
Craft reusable prompt templates with quality gates and reasoning guidance. Orchestrate multi-step workflow chains with a composable operator syntax. Export as native skills to any client.
Quick Start · What You Get · Compose Workflows · Run Anywhere · Docs
What your AI client gives you — and what this server adds
| Your client already does | This server adds |
|---|---|
| Run a prompt | Compose prompts with validation, reasoning guidance, and formatting in one expression |
| Single-shot skills | Multi-step workflows that thread context between steps |
| Execute subagents | Hand off mid-chain steps to agents with full workflow context |
| Client-native skill format | Author once as YAML, export to any client with skills:export |
| Manual prompt writing | Versioned templates with hot-reload, rollback, and history |
| Trust the output | Validate output between steps: self-evaluation and shell commands |
Is this for me?
- Use this if you write the same prompts repeatedly, run multi-step workflows, or want to share reusable prompts with a team.
- Skip if your client's built-in
/commandsalready handle what you need, or you're looking for a no-code prompt library. - Works with Claude Code, Claude Desktop, Cursor, OpenCode, Gemini CLI, Codex, Windsurf, and Zed. Plugin installers add hooks (chain tracking, gate enforcement, state preservation) for Claude Code, OpenCode, Gemini CLI, and Codex (experimental); other clients run MCP-only.
Quick Start
Claude Code (Recommended)
# Add marketplace (first time only)
/plugin marketplace add minipuft/minipuft-plugins
# Install
/plugin install claude-prompts@minipuft
# Try it
>>tech_evaluation_chain library:'zod' context:'API validation'
Load plugin from local source for development:
git clone https://github.com/minipuft/claude-prompts-mcp ~/Applications/claude-prompts-mcp
cd ~/Applications/claude-prompts-mcp/server && npm install && npm run build
claude --plugin-dir ~/Applications/claude-prompts-mcp
Edit hooks/prompts → restart Claude Code. Edit TypeScript → rebuild first.
Claude Desktop
Option A: GitHub Release (recommended)
- Download
claude-prompts-{version}.mcpbfrom Releases - Drag into Claude Desktop Settings → MCP Servers
- Done
The .mcpb bundle is self-contained (~5MB); no npm required.
Option B: NPX (auto-updates)
Add to your config file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"claude-prompts": {
"command": "npx",
"args": ["-y", "claude-prompts@latest", "--client", "claude-code"]
}
}
}
Restart Claude Desktop and test: >>research_chain topic:'remote team policies'
One-click install: the VS Code and Cursor buttons at the top of this README register the server via npx.
Plugin installers (recommended, adds hooks):
# OpenCode (full hooks)
npm install -g opencode-prompts && opencode-prompts install
# Gemini CLI (partial hooks)
gemini extensions install https://github.com/minipuft/gemini-prompts
# Codex (experimental hooks; requires hooks = true under [features] in ~/.codex/config.toml)
codex plugin marketplace add https://github.com/minipuft/minipuft-plugins.git
codex plugin add codex-prompts@minipuft
Manual config for VS Code, Cursor, OpenCode (no hooks), Gemini CLI (no hooks), Codex, Windsurf, Zed: see Client Integration Guide for per-client config locations, JSON examples, and --client preset matrix. Client Capabilities Reference covers profile mapping and limits.
From source (developers):
git clone https://github.com/minipuft/claude-prompts-mcp.git
cd claude-prompts-mcp/server && npm install && npm run build && npm test
Point your MCP config to server/dist/index.js. Transport: --transport=stdio (default) or --transport=streamable-http.
Custom resources: --init=~/my-prompts scaffolds a starter workspace: three example prompts plus config.json. Edit them (YAML schema), or have your AI author new prompts, gates, and frameworks via resource_manager. Point MCP_RESOURCES_PATH at an existing workspace if you already have one in the right shape. See Custom Resources Guide.
What You Get
Four primitives you author, version, and compose. The bundled set ships 120+ prompts across 17 categories. All hot-reloadable, all versioned with rollback.
| Primitive | Symbol | What it is | Example |
|---|---|---|---|
| Prompt template | >> | Versioned YAML with named arguments; hot-reload on save | >>code_review target:'src/auth/' |
| Gate | :: | Validation criterion the AI checks its own output against; blocking or advisory; can shell-verify | :: 'cite sources' · :: verify:"npm test" |
| Framework | @ | Reasoning framework that shapes how the AI works through the problem; plug in your own or use built-ins like @ReACT, @5W1H, or the project's own @CAGEERF scaffold (Frameworks Guide) | @ReACT · @your_framework |
| Style | # | Output formatting and tone | #analytical · #procedural |
Prompts, gates, and frameworks are managed through the resource_manager tool. Your AI creates, edits, versions, and rolls them back through MCP, no file editing required. Styles are managed with the bundled cpm CLI. Failed gate checks can retry automatically or pause for your decision (Gates Guide). Build your first primitive: Prompt Authoring Tutorial.
The Three Tools
Everything above reaches your client through three MCP tools:
| Tool | Purpose |
|---|---|
prompt_engine | Execute prompts with frameworks and validation |
resource_manager | Create, update, version, and roll back resources |
system_control | Status, analytics, framework switching |
Most users invoke these via >> syntax in conversation; hooks construct the actual calls. For programmatic MCP clients calling tools directly, see MCP Tools Reference.
Compose Workflows
How to write a chain
>>review target:'src/auth/' @ReACT :: 'cite sources'
--> security_scan :: verify:"npm test"
==> implementation
Read top-to-bottom:
>>review target:'src/auth/'runs thereviewprompt against your auth folder.@ReACToverlays the ReACT reasoning framework on this step.:: 'cite sources'adds a gate the AI must satisfy (cite sources, or retry).--> security_scan :: verify:"npm test"chains to step 2, which must passnpm testbefore producing output.==> implementationhands the final step off to a client-native agent (a subagent in Claude Code).
Validation runs between steps, not only at the end. For the full operator set including *, %, and #, see the Syntax Reference details block near the bottom.
A gate catches a missing field, the model corrects itself, and the chain passes. Recorded on haiku, the cheapest model.
Two patterns extend the basic syntax. Chains also support context threading between steps and agent handoffs. See Chains Lifecycle and MCP Tools Reference.
Context7 fetches live library docs mid-chain. The final output is a structured assessment with sources.
Verification Loops
Ground-truth validation via shell commands. The AI keeps iterating until tests pass:
>>implement-feature :: verify:"npm test" loop:true
Implements, runs the test, reads failures, fixes, retries. Spawns a fresh context after repeated failures to avoid context rot.
| Preset | Tries | Timeout | Use Case |
|---|---|---|---|
:fast | 1 | 30s | Quick check |
:full | 5 | 5 min | CI validation |
:extended | 10 | 10 min | Large test suites |
For autonomous test-fix cycles with context-rot prevention: Ralph Loops Guide.
Judge Mode
Let the AI pick the right resources for the task:
%judge Help me refactor this authentication module
Analyzes available templates, reasoning frameworks, validation rules, and styles, then recommends the best combination. You confirm before it runs. For scoring and overrides see Judge Mode Guide.
Run Anywhere
Author workflows as YAML templates. Export as native skills to your client.
# skills-sync.yaml — choose what to export
registrations:
claude-code:
user:
- prompt:development/review
- prompt:development/validate_work
npm run skills:export
The review prompt becomes a /review Claude Code skill. validate_work becomes /validate_work. Same source, native experience; no MCP call required at runtime.
Compiles to Claude Code skills, Cursor rules, OpenCode commands, and more. npm run skills:diff flags when exports drift from source. Configuration, supported clients, and drift detection: Skills Sync Guide.
With Hooks
Without hooks, you're calling the three MCP tools explicitly (the LLM constructs each call). With hooks, the operators work in conversation: >>, -->, ==>, :: feel native rather than mediated, and workflow state survives across LLM turns and context compaction.
What hooks unlock:
| Hook | Unlocks |
|---|---|
| Auto-routing | >>analyze topic:'X' in chat fires the right MCP tool call without you naming it |
| Chain continuity across compaction | Multi-step chains preserve state when context compacts mid-execution; the chain doesn't restart from scratch |
| Cross-step verdict tracking | Gate pass/fail verdicts thread across all chain steps without the LLM re-deriving them |
| Native agent handoffs | ==> routes to your client's subagent system automatically; no manual subagent invocation |
| Session persistence | Workflow state preserved when context compacts mid-chain |
Hooks ship with the plugin install. Full support on Claude Code (this repo) and OpenCode; partial on Gemini CLI; experimental on Codex, where Codex hooks are off by default and each install requires a one-time /hooks trust review. Other clients get the three MCP tools but no hook-driven behaviors. Detail: hooks/README.md.
| Symbol | Name | What It Does | Example |
|---|---|---|---|
>> | Prompt | Execute template | >>code_review |
--> | Chain | Pipe to next step | step1 --> step2 |
==> | Handoff | Route step to agent | step1 ==> agent_step |
* | Repeat | Run prompt N times | >>brainstorm * 5 |
@ | Framework | Inject reasoning guidance | @ReACT |
:: | Gate | Add validation criteria | :: 'cite sources' |
% | Modifier | Toggle behavior | %clean, %judge |
# | Style | Apply formatting | #analytical |
Modifiers (one per command, placed first):
%clean— No framework/gate/style injection%lean— Gates only, skip framework and style%framework— Framework injection only%judge— AI recommends resources; you confirm
→ MCP Tools Reference for full command documentation.
How It Works
Command with operators → server parses and injects resources (framework, gates, style) → client executes the rendered prompt and self-evaluates against the gates → router decides: next step on pass, retry on fail, return on done.
Full request lifecycle, pipeline stages, and subsystem diagrams: Architecture Overview.
Documentation
Full docs index, organized by Diátaxis quadrant (tutorials, how-to, reference, concepts):
Quick jumps: Build your first prompt · Chains lifecycle · MCP Tools reference · Architecture overview · Troubleshooting
Contributing
cd server
npm install
npm run build # esbuild bundles to dist/index.js
npm test # Run test suite
npm run validate:all # Full CI validation
The build produces a self-contained bundle. server/dist/ is gitignored, and CI builds fresh from source.
See CONTRIBUTING.md for workflow details.
License
File tree
1100minipuft/claude-prompts-mcpView on GitHub
.claude/
.claude-plugin/
plugin.json
rules/
extension-alignment.md
mcp-contracts.md
sqlite-persistence.md
.editorconfig
.gitattributes
.github/
actions/
build-server/
action.yml
extract-version/
action.yml
setup-node-install/
action.yml
validate-required-files/
action.yml
ISSUE_TEMPLATE/
bug_report.yml
config.yml
feature_request.yml
pull_request_template.md
renovate.json5
required-contexts.json
workflows/
ci.yml
downstream-sync.yml
extension-publish.yml
npm-publish.yml
registry-publish.yml
release-please.yml
renovate-config-validator.yml
.gitignore
.husky/
commit-msg
pre-commit
pre-push
.mcp.json
.mcpbignore
.node-version
.prettierignore
.release-please-manifest.json
agents/
chain-executor.md
assets/
demos/
chain-workflow-demo.gif
hero-chain-demo.gif
hero-demo.gif
resource-list-demo.gif
skills-export-demo.gif
skills-export.gif
skills-export.tape
status-demo-3x.gif
logo.png
CHANGELOG.md
CLAUDE.md
cli/
esbuild-plugins/
zod-locales-trim.mjs
esbuild.config.mjs
jest.config.cjs
package.json
README.md
src/
cli.ts
commands/
compare.ts
config.ts
create.ts
delete.ts
enable-disable.ts
guide.ts
history.ts
init.ts
inspect.ts
link-gate.ts
list.ts
move.ts
rename.ts
rollback.ts
toggle.ts
validate.ts
index.ts
lib/
output.ts
resource-validation.ts
types.ts
workspace.ts
tests/
fixtures/
empty-workspace/
resources/
.gitkeep
invalid-workspace/
resources/
prompts/
dev/
broken/
prompt.yaml
valid-workspace/
resources/
frameworks/
test-method/
framework.yaml
gates/
test-gate/
gate.yaml
guidance.md
prompts/
dev/
hello/
prompt.yaml
user-message.md
versioned-workspace/
resources/
frameworks/
test-method/
framework.yaml
gates/
test-gate/
gate.yaml
guidance.md
prompts/
general/
test-prompt/
prompt.yaml
user-message.md
styles/
test-style/
guidance.md
style.yaml
integration/
cli.test.ts
new-commands.test.ts
structural-commands.test.ts
tsconfig.json
commitlint.config.mjs
CONTRIBUTING.md
docs/
adr/
0000-template.md
0001-gate-resolution-precedence.md
README.md
architecture/
overview.md
concepts/
chains-lifecycle.md
quality-gates.md
guides/
chain-authoring-example.md
cli.md
client-integration.md
custom-resources.md
frameworks.md
gates.md
identity-scope.md
injection-control.md
judge-mode.md
phase-guards.md
ralph-loops.md
release-process.md
script-tools.md
skills-sync.md
telemetry-observability.md
troubleshooting.md
workspace-organization-identity-migration.md
how-to/
add-validation.md
portfolio/
case-study.md
design-decisions.md
readme-charter.md
README.md
reference/
chain-schema.md
client-capabilities.md
gate-configuration.md
mcp-tools.md
otel-collector-tail-sampling.yaml
prompt-yaml-schema.md
template-syntax.md
TODO.md
tutorials/
build-first-prompt.md
glama.json
hooks/
.pyrefly-baseline.json
compact-recovery.py
delegation-enforce.py
gate-enforce.py
hooks.json
lib/
cache_manager.py
cli_spawner.py
config_loader.py
db_reader.py
hook_state_store.py
lesson_extractor.py
model_strategies.py
operators.py
ralph_subagent_contract.py
session_state.py
session_tracker.py
task_protocol.py
verify_active_store.py
workspace.py
post-prompt-engine.py
prompt-suggest.py
ralph-context-tracker.py
ralph-stop.py
README.md
subagent-gate-enforce.py
tests/
conftest.py
test_codex_client_seams.py
test_criterion_coverage.py
test_db_reader.py
test_gate_enforce_verdict.py
test_integration_ralph_delegation.py
test_operator_detection.py
test_ralph_stop.py
test_ralph_subagent_contract.py
test_session_tracker.py
test_subagent_gate_enforce.py
test_task_protocol.py
test_verify_state_schema_parity.py
LICENSE
manifest.json
package-lock.json
package.json
plans/
acquisition-recovery-implementation-notes.md
acquisition-recovery.md
claude-prompts-mcp.code-workspace
cli-distribution-release-integration-2026-08-02-implementation-notes.md
cli-distribution-release-integration-2026-08-02.md
codex-prompts-port-2026-08-03.md
codex-prompts-port-implementation-notes.md
dashboard-refactor.md
demo-video-plan.md
downstream-standards-federation-2026-08-02-implementation-notes.md
downstream-standards-federation-2026-08-02.md
features/
plan-retirement-federation-2026-08-03.md
subagent-selection-2026-08-03.md
mcp-sdk-v2-spec-2026-07-28-migration.md
readme-restructure.md
reference/
semantic-llm-sidecar-retirement-2026-08-05-implementation-notes.md
semantic-llm-sidecar-retirement-2026-08-05.md
sqlite-layer-remediation-2026-08-03-implementation-notes.md
sqlite-layer-remediation-2026-08-03.md
techincal_debt/
arg-gate-pipeline-fixes.md
compat-site-audit.md
implementation-notes.md
pipeline-defect-remediation-2026-08-01.md
pipeline-followup-2026-08-02.md
pipeline-followup-2026-08-03.md
reader-without-producer-sweep-2026-08-05-implementation-notes.md
reader-without-producer-sweep-2026-08-05.md
stepstate-lifecycle-migration.md
type_safety.md
renovate-maintenance-remediation-2026-08-01-implementation-notes.md
renovate-maintenance-remediation-2026-08-01.md
techincal_debt/
shim-debt-sweep-2026-07-29.md
test-modernization-roadmap.md
validation-mechanism-architecture-2026-08-05-implementation-notes.md
validation-mechanism-architecture-2026-08-05.md
validation-surface-cleanup-2026-08-04.md
project-decisions.md
pyproject.toml
pyrefly.toml
README.md
release-please-config.json
ruff.toml
scripts/
build-extension.sh
classify-validation-scope.js
dev-claude.sh
merge-unreleased-changelog.js
retire-done-plans.js
stage-server-runtime.sh
sync-to-cache.sh
validate-hook-registration.js
server/
server.json
.dependency-cruiser.cjs
.eslint-ratchet-baseline.json
.npmignore
.npmrc
.prettierignore
.prettierrc.json
.typecheck-tests-ratchet-baseline.json
config.json
config.schema.json
esbuild.config.mjs
eslint-rules/
claude-plugin.d.ts
claude-plugin.js
eslint.config.js
jest.config.cjs
knip.json
LICENSE
package-lock.json
package.json
README.md
resources/
frameworks/
5w1h/
framework.yaml
judge-prompt.md
phases.yaml
cageerf/
framework.yaml
judge-prompt.md
phases.yaml
focus/
framework.yaml
judge-prompt.md
phases.yaml
system-prompt.md
liquescent/
framework.yaml
judge-prompt.md
phases.yaml
system-prompt.md
radiant/
framework.yaml
judge-prompt.md
phases.yaml
react/
framework.yaml
judge-prompt.md
phases.yaml
scamper/
framework.yaml
judge-prompt.md
phases.yaml
verify/
framework.yaml
judge-prompt.md
phases.yaml
system-prompt.md
gates/
_index.md
api-documentation/
gate.yaml
guidance.md
code-quality/
gate.yaml
guidance.md
config/
shell-presets.yaml
verdict-patterns.yaml
content-structure/
gate.yaml
guidance.md
creed-fidelity/
gate.yaml
guidance.md
educational-clarity/
gate.yaml
guidance.md
framework-compliance/
gate.yaml
guidance.md
intent-quality/
gate.yaml
guidance.md
math-fidelity/
gate.yaml
guidance.md
plan-quality/
gate.yaml
guidance.md
pr-performance/
gate.yaml
guidance.md
pr-security/
gate.yaml
guidance.md
research-quality/
gate.yaml
guidance.md
security-awareness/
gate.yaml
guidance.md
technical-accuracy/
gate.yaml
guidance.md
test-coverage/
gate.yaml
guidance.md
test-suite/
gate.yaml
... 1046 moreFAQ
claude-prompts is a Claude Code plugin with hand-picked skills for automation work, indexed on Flowy. Install it with the command on its page. Its skills do not fire on their own yet. Request auto-invocation to have Flowy route them as you prompt. Free and open source.
- What is claude-prompts?
- A Model Context Protocol (MCP) server for prompt workflows. Written once, always followed. Craft reusable prompt templates with quality gates and reasoning guidance. Orchestrate multi-step workflow chains with a composable operator syntax.
- How do I install claude-prompts?
- Run these in Claude Code: . Then prompt normally.
- Does claude-prompts auto-invoke its skills?
- Not yet. It is indexed on Flowy as a plain plugin. Request auto-invocation on its page and Flowy will route its skills for you as you prompt.
- Is claude-prompts free?
- Yes. Flowy is free and open source, with nothing gated. You can read every skill in full before you install.