By zazza123
Universal Python library documentation for AI coding agents via the Library Context Protocol. Resolves any pip-installed package on demand — functions, classes, signatures, docstrings — through browsable MCP tools.
This skill should be used when the user writes code that imports or uses a Python library, asks to "look up the X API", "check how to use X", "what's the signature of X.Y", "resolve library X", or encounters import errors or API misuse. Activates the lcp MCP server's resolve_library workflow for on-demand introspection of any pip-installed package.
This skill should be used when the user writes code against a Python library, asks "how do I use X", encounters an AttributeError or ImportError on a library symbol, or asks "what parameters does X.Y take". Detects whether the universal lcp server or a per-library lcp-LIBNAME server is connected and guides the agent to use the correct exploration workflow.
Admin access level
Server config contains admin-level keywords
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
This plugin requires configuration values that are prompted when the plugin is enabled. Sensitive values are stored in your system keychain.
registriesComma-separated list of LCP registry URLs. The first URL is used as the primary registry for fetching pre-built manifests. Leave empty to use only the official registry.
${user_config.registries}
Documentation: https://zazza123.github.io/lcp
lcp (Library Context Protocol) is primarly a protocol designed to solve the problem of AI agents not having access to up-to-date library documentation, which leads to hallucinations and inaccurate code generation. The LCP SDK provides tools to scan Python packages, extract API information, and generate LCP-compliant JSON manifests. It also includes features for analyzing documentation coverage and generating missing docstrings using AI.
pip install lcp
inspect and ast moduleslcp[ai])# Scan a package and output LCP JSON
lcp scan requests -o requests.lcp.json
# Include private symbols
lcp scan mypackage --include-private
# Skip validation
lcp scan mypackage --no-validate
# Start an MCP server for a library manifest
lcp serve requests.lcp.json
from lcp import scan
# Scan a package
lcp_doc = scan("requests")
# Save to file
lcp_doc.to_file("requests.lcp.json")
# Get as dict
data = lcp_doc.to_dict()
# Include private symbols
lcp_doc = scan("mypackage", include_private=True)
Analyze documentation completeness of a package to identify missing docstrings.
# Generate coverage report (JSON)
lcp coverage requests -o coverage.json
# Generate coverage report (Markdown)
lcp coverage requests -o coverage.md --format markdown
# Generate both LCP manifest and coverage report in one scan
lcp scan requests -o requests.lcp.json --coverage coverage.json
from lcp import generate_coverage
# Generate coverage report
report = generate_coverage("requests")
# Check coverage percentage
print(f"Coverage: {report.summary.coverage_percent}%")
print(f"Documented: {report.summary.documented}/{report.summary.total_symbols}")
# List undocumented symbols
for symbol in report.undocumented:
print(f" - {symbol.module}:{symbol.entity} ({symbol.kind})")
# Save report
report.to_file("coverage.json") # JSON format
report.to_file("coverage.md") # Markdown format
Compare two LCP manifests to detect symbols that were removed between versions and automatically generate deprecation entries.
# Compare two versions and print the diff report
lcp diff v1.lcp.json v2.lcp.json
# Save the diff report to a file
lcp diff v1.lcp.json v2.lcp.json -o diff.json
# Automatically update the new manifest with deprecation entries
lcp diff v1.lcp.json v2.lcp.json --update
from lcp import diff_documents, load_lcp_document, update_document
# Load two versions
old = load_lcp_document("v1.lcp.json")
new = load_lcp_document("v2.lcp.json")
# Compare
result = diff_documents(old, new)
print(f"Removed: {len(result.removed)}, Added: {len(result.added)}")
for sid, dep in result.deprecated.items():
print(f" {sid}: deprecated in {dep.deprecated_in}")
# Merge deprecations into the new document
updated = update_document(new, result)
updated.to_file("v2.lcp.json")
The SDK includes an MCP (Model Context Protocol) server that exposes LCP manifest data to AI agents. This allows agents to explore library APIs and generate accurate code.
# Start MCP server for a library
lcp serve requests.lcp.json
# With custom server name
lcp serve numpy.lcp.json --name numpy-docs
Add to your MCP client configuration (e.g., Claude Desktop):
{
"mcpServers": {
"requests-api": {
"command": "lcp",
"args": ["serve", "/path/to/requests.lcp.json"]
}
}
}
| Tool | Description |
|---|---|
get_manifest | Get library metadata (name, version, language) |
list_modules | List all modules in the library |
list_symbols | Browse symbols with optional filtering by module or kind |
get_symbol | Get full details for a specific symbol |
search_symbols | Find symbols by text search |
get_class_members | Get all methods and attributes of a class |
npx claudepluginhub zazza123/lcp --plugin lcpHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
Comprehensive C4 architecture documentation workflow with bottom-up code analysis, component synthesis, container mapping, and context diagram generation
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review