{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_ynqdjuu4mjbn","handle":"terse-mcp","url":"https://wellknown.network/agents/terse-mcp","links":{"self":"https://wellknown.network/agents/terse-mcp/record.json","html":"https://wellknown.network/agents/terse-mcp","markdown":"https://wellknown.network/agents/terse-mcp/record.md","api":"https://wellknown.network/api/v1/agents/terse-mcp","status":"https://wellknown.network/api/v1/agents/terse-mcp/status","claim":"https://wellknown.network/agents/terse-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/terse-mcp/claim.json","badge":"https://wellknown.network/agents/terse-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/terse-mcp/history","tools":"https://wellknown.network/api/v1/agents/terse-mcp/tools"},"ard":{"identifier":"urn:air::server:terse-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"terse-mcp","summary":"Lossless-first compression layer for AI-agent tool outputs — byte-faithful on fields you mark critical, configurable lossy reduction only where you opt in.","description":"# terse\n\n[![tests](https://github.com/inth3shadows/terse/actions/workflows/tests.yml/badge.svg)](https://github.com/inth3shadows/terse/actions/workflows/tests.yml)\n[![PyPI](https://img.shields.io/pypi/v/terse-mcp.svg)](https://pypi.org/project/terse-mcp/)\n[![Python](https://img.shields.io/badge/python-3.11%2B-blue.svg)](pyproject.toml)\n[![License: MIT](https://img.shields.io/badge/license-MIT-green.svg)](LICENSE)\n[![terse MCP server](https://glama.ai/mcp/servers/inth3shadows/terse/badges/score.svg)](https://glama.ai/mcp/servers/inth3shadows/terse)\n\n**Cut the tokens your AI agent spends reading tool output, without changing a single value it reads.**\nterse is a drop-in proxy for any MCP server. It strips structural overhead from each\nresult (repeated keys, repeated values, repeated nested shapes) and hands the model the\nsame data, value for value, in a denser form. No decode step, no model in the loop.\n\nFive GitHub-style issues cost 436 tokens (cl100k) as a server pretty-prints them, and 232\nminified. What the model gets instead costs 181:\n\n```json\n{\"__terse_dict__\":1,\"legend\":{\"~0\":[{\"name\":\"bug\",\"color\":\"d73a4a\"}]},\n \"data\":{\"__terse_table__\":1,\"n\":5,\"cols\":[\"number\",\"state\",\"title\",\"user\",\"labels\",\"repo\"],\n  \"rows\":[[101,\"open\",\"Crash on empty cart\",[\"ana\",\"User\"],\"~0\",\"acme/widgets\"],\n          [102,\"open\",\"Timeout on checkout\",[\"ben\",\"User\"],\"~0\",\"acme/widgets\"], ...],\n  \"subcols\":{\"user\":{\"cols\":[\"login\",\"type\"]}}}}\n```\n\n- **Lossless, and checked.** Every transform has an exact inverse, and a round-trip gate\n  asserts `decompress(compress(x)) == x`. Other tools in this space drop rows or old\n  results to save tokens; terse only removes repetition.\n- **Measured against the alternatives.** On real GitHub API payloads: **59.1%** fewer\n  tokens, against −6.5% for TOON and 4.5% for compressmcp on the same corpus\n  ([Benchmarks](#benchmarks-terse-vs-alternatives)).\n- **Selective, never worse.** It works per tool and passes through anything it can't\n  shrink, so it n…","publisher":{"name":"inth3shadows","url":"https://github.com/inth3shadows"},"homepage":"https://github.com/inth3shadows/terse","repository":"https://github.com/inth3shadows/terse","version":"0.49.0","license":null,"protocols":["mcp"],"tags":["agent","compression","context-window","json","llm","lossless","mcp","proxy","tokens","python","macos","windows","linux","aggregators"],"pricing":null,"endpoints":[{"url":"pypi:terse-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":{"section":"Aggregators","listedIn":"punkpeye/awesome-mcp-servers","deployment":["local"]},"attribution":{"kind":"pypi","name":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi","publisherUrl":"github","publisherName":"github"}},"derived":{"capabilities":[{"slug":"dev.version-control","name":"Version Control","confidence":0.779,"provenance":"derived"}],"categories":["dev"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"terse-mcp","registry":"pypi","observedAt":"2026-11-04T02:26:22.850Z","publishedAt":"2026-10-30T00:23:24.830616Z","latestVersion":"0.49.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"github","key":"github.com/inth3shadows/terse","url":"https://github.com/inth3shadows/terse","firstSeenAt":"2026-10-18T00:22:40.531Z","fetchedAt":"2026-11-05T04:18:54.283Z","normalizedAt":"2026-11-05T04:18:54.283Z"},{"source":"pypi","key":"terse-mcp","url":"https://pypi.org/project/terse-mcp/","firstSeenAt":"2026-09-10T14:22:01.750Z","fetchedAt":"2026-11-04T02:24:53.435Z","normalizedAt":"2026-11-04T02:24:53.435Z"}]},"firstSeenAt":"2026-09-10T14:22:01.750Z","updatedAt":"2026-11-04T02:26:22.850Z"}