TASS MCP Server
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
TASS Community Edition
Text Analysis for Social Scientists: LIWC-style dictionary scoring over open lexicons: deterministic, provenance-first, driveable by humans (CLI) and by AI agents (MCP server). Apache-2.0, zero runtime dependencies, Node 18+.
TASS is built by SIM DAD LLC. This repository is the full
Community Edition: the scoring engine, the tass CLI, the MCP server, the minimal
local web GUI, and the reproducible-project container. The commercial GUI Edition
adds a no-code desktop workbench, the inferential statistics suite, publication
charts, and Excel/APA export on top of this exact engine; its methods, formulas,
citations, and R-benchmark reports are public even where its code is not.
Why researchers use it
Deterministic. Same inputs, byte-identical outputs, on every OS. CI enforces a cross-platform determinism hash. Reproducibility is a product claim, not a hope.
Provenance travels. Every scoring run writes a manifest with tool and engine versions, input SHA-256 hashes, and each lexicon's license, class, and citation: the block a methods section needs.
Licensing is mechanical. Everything bundled is verified commercial-ok. Restricted resources (NRC, LIWC, GI) are never bundled; your own imports of them are flagged academic-only in every run that touches them.
Trace every number.
tass exemplarsandtass kwicconnect any score back to quotable text.Open dictionaries, forever. Every TASS-native dictionary is CC-BY-4.0 or ODC-BY, citation required, commercial use welcome, and installable by name from the public registry (SIM-DAD/tass-lexicons).
Related MCP server: mhlabs-mcp-tools
Quick start
The fastest route is npm (Node 18+):
npm install -g @simdad/tass-cli
tass helpOr from this repository:
npm install && npm run build
node packages/tass-cli/bin/tass.js helpTo put the repo build on your PATH: on Windows run
powershell -File packages/tass-cli/scripts/install-cli.ps1; on macOS or Linux run
npm link inside packages/tass-cli. Then:
Score a CSV of documents against bundled sentiment lexicons:
tass score --input posts.csv --text-column text --lexicons vader,afinn --output scored.csvAnalyze a single text, no files involved:
tass analyze --text "What a wonderful day" --vader-rulesAuthor your own dictionary in a spreadsheet (tass template, fill it in, tass import-csv), install open ones by name (tass install politeness), archive any
run as a reproducible .tassproj (tass project save), and prove it reproduces
(tass project rerun).
For AI-driven analysis, tass mcp serves the whole engine as MCP tools over stdio.
AGENTS.md is the operating contract for agents.
Documentation
docs/METHODS.md: what TASS measures and how; read before reporting numbers.docs/API.md: the engine's semver contract.tass help <topic>: task-oriented help inside the CLI.
Contributing
See CONTRIBUTING.md (Apache-2.0, DCO sign-off, determinism and zero-dependency
rules). Dictionary contributions go to the registry repo.
Apache-2.0 © SIM DAD LLC. "TASS" is a SIM DAD LLC trademark; see NOTICE.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Structured analysis API and remote MCP tool for text, JSON records and numeric series.
MCP-native AI evaluation: rubric audits, eval suites, and proof reports for AI/LLM output.
Provenance and metadata tools for AI agents. Paid tools use x402 (USDC) at /mcp/<tool>.
Free OpenAI-compatible inference with signed provenance receipts and 3 focused MCP tools.
Related MCP Servers
- AlicenseNot gradedqualityNot gradedmaintenanceProvides text analysis tools including TF-IDF document search and text profiling (sentiment, readability, keywords) for analyzing text corpora through structured MCP interfaces.BSD 3-Clause
- AlicenseNot gradedqualityDmaintenanceModular MCP server providing text preprocessing and NLP tools for AI agent ecosystems.MIT
- AlicenseNot gradedqualityDmaintenanceAlgorithmic text and NLP analysis server providing sentiment, readability, keyword extraction, language detection, stats, and more via REST and MCP.11 npmMIT
- FlicenseNot gradedqualityAmaintenanceEnables agents to analyze text or files for formulaic AI writing patterns and receive numeric scores, band labels, contextual violations, and concrete advice. It offers programmatic prose linting via MCP tools without using an LLM judge or external API calls.-