chart-library-mcp/smithery.yaml at master · grahammccain/chart-library-mcp · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
# Smithery server config — https://smithery.ai
#
# Chart Library MCP — chart-pattern intelligence for agents.
# 14 canonical tools backed by 25M+ embeddings, 19K+ symbols,
# 10 years of US equity history. Output is a calibrated
# distribution of historical analogs, not a forecast.

name: chart-library
version: 6.1.0
description: >
  Chart-pattern intelligence for AI agents. Flagship tool pull_comps
  pulls the comp set for a subject (symbol, date, timeframe); 14
  canonical tools in all (search, pull_comps, cohort_introspect,
  symbol_intelligence, cohort_members, cohort_groupby, cohort_rerank,
  cohort_attribution, track_record, analyze, context, explain,
  portfolio, report_feedback) backed by 25M+ embeddings, 19K+ symbols,
  10y US equity history. The core loop is search → pull_comps →
  cohort_introspect: anchor a (symbol, date, timeframe), get the comp
  set of historical analogs, then slice it. Calibrated distributions,
  not forecasts.

homepage: https://chartlibrary.io
repository: https://github.com/grahammccain/chart-library-mcp

# Local stdio install path — Smithery installs from PyPI
# (already published as `chartlibrary-mcp`) and launches the
# `chartlibrary-mcp` console script entry point.
startCommand:
  type: stdio
  configSchema:
    type: object
    properties:
      CHART_LIBRARY_API_KEY:
        type: string
        title: API key
        description: >
          Free Sandbox key gives 200 calls/day on search +
          context. Get one at https://chartlibrary.io/developers.
          Optional — the server runs without a key but deeper
          tools (cohort_analyze, cohort_introspect, explain)
          require auth.
  commandFunction: |-
    (config) => ({
      command: "chartlibrary-mcp",
      args: [],
      env: {
        CHART_LIBRARY_API_KEY: config.CHART_LIBRARY_API_KEY || ""
      }
    })

# Tool surface — flagship pull_comps + the canonical-14. cohort_analyze
# is the same engine under the original field names (kept callable but
# hidden from tools/list so pull_comps is the one discoverable flagship);
# deprecated umbrella wrappers (cohort, market_briefing, anchor_fetch,
# narrative_*, discover_picks) stay callable but are not advertised here.
# Read-only tools auto-allow in safe-mode MCP clients. The core loop is
# search → pull_comps → cohort_introspect for any (symbol, date) anchor.
tools:
  - name: search
    annotation: read_only
  - name: pull_comps
    annotation: read_only
  - name: cohort_introspect
    annotation: read_only
  - name: symbol_intelligence
    annotation: read_only
  - name: cohort_members
    annotation: read_only
  - name: cohort_groupby
    annotation: read_only
  - name: cohort_rerank
    annotation: read_only
  - name: cohort_attribution
    annotation: read_only
  - name: track_record
    annotation: read_only
  - name: analyze
    annotation: read_only
  - name: context
    annotation: read_only
  - name: explain
    annotation: read_only
  - name: portfolio
    annotation: read_only
  - name: report_feedback
    annotation: write