@speakeasy-api/docs-mcp-core 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/LICENSE +661 -0
- package/README.md +27 -0
- package/dist/chunking.d.ts +3 -0
- package/dist/chunking.d.ts.map +1 -0
- package/dist/chunking.js +332 -0
- package/dist/chunking.js.map +1 -0
- package/dist/cursor.d.ts +12 -0
- package/dist/cursor.d.ts.map +1 -0
- package/dist/cursor.js +83 -0
- package/dist/cursor.js.map +1 -0
- package/dist/embedding-cache.d.ts +46 -0
- package/dist/embedding-cache.d.ts.map +1 -0
- package/dist/embedding-cache.js +230 -0
- package/dist/embedding-cache.js.map +1 -0
- package/dist/embedding.d.ts +66 -0
- package/dist/embedding.d.ts.map +1 -0
- package/dist/embedding.js +264 -0
- package/dist/embedding.js.map +1 -0
- package/dist/index.d.ts +11 -0
- package/dist/index.d.ts.map +1 -0
- package/dist/index.js +11 -0
- package/dist/index.js.map +1 -0
- package/dist/lancedb.d.ts +42 -0
- package/dist/lancedb.d.ts.map +1 -0
- package/dist/lancedb.js +442 -0
- package/dist/lancedb.js.map +1 -0
- package/dist/manifest.d.ts +19 -0
- package/dist/manifest.d.ts.map +1 -0
- package/dist/manifest.js +245 -0
- package/dist/manifest.js.map +1 -0
- package/dist/metadata.d.ts +6 -0
- package/dist/metadata.d.ts.map +1 -0
- package/dist/metadata.js +152 -0
- package/dist/metadata.js.map +1 -0
- package/dist/parser.d.ts +6 -0
- package/dist/parser.d.ts.map +1 -0
- package/dist/parser.js +21 -0
- package/dist/parser.js.map +1 -0
- package/dist/search-common.d.ts +7 -0
- package/dist/search-common.d.ts.map +1 -0
- package/dist/search-common.js +81 -0
- package/dist/search-common.js.map +1 -0
- package/dist/search.d.ts +13 -0
- package/dist/search.d.ts.map +1 -0
- package/dist/search.js +166 -0
- package/dist/search.js.map +1 -0
- package/dist/types.d.ts +125 -0
- package/dist/types.d.ts.map +1 -0
- package/dist/types.js +2 -0
- package/dist/types.js.map +1 -0
- package/package.json +49 -0
package/dist/search.js
ADDED
|
@@ -0,0 +1,166 @@
|
|
|
1
|
+
import { decodeSearchCursor, encodeSearchCursor } from "./cursor.js";
|
|
2
|
+
import { clampLimit, isChunkIdFormat, makeSnippet, matchesMetadataFilters, normalizeSearchText, tokenizeSearchText } from "./search-common.js";
|
|
3
|
+
export class InMemorySearchEngine {
|
|
4
|
+
chunks;
|
|
5
|
+
byId;
|
|
6
|
+
byFile;
|
|
7
|
+
proximityWeight;
|
|
8
|
+
constructor(chunks, options = {}) {
|
|
9
|
+
this.chunks = [...chunks];
|
|
10
|
+
this.byId = new Map(chunks.map((chunk) => [chunk.chunk_id, chunk]));
|
|
11
|
+
this.byFile = new Map();
|
|
12
|
+
this.proximityWeight = options.proximityWeight ?? 1.25;
|
|
13
|
+
for (const chunk of chunks) {
|
|
14
|
+
const list = this.byFile.get(chunk.filepath) ?? [];
|
|
15
|
+
list.push(chunk);
|
|
16
|
+
this.byFile.set(chunk.filepath, list);
|
|
17
|
+
}
|
|
18
|
+
for (const list of this.byFile.values()) {
|
|
19
|
+
list.sort((a, b) => a.chunk_index - b.chunk_index);
|
|
20
|
+
}
|
|
21
|
+
}
|
|
22
|
+
async search(request) {
|
|
23
|
+
const query = request.query.trim();
|
|
24
|
+
if (!query) {
|
|
25
|
+
throw new Error("query is required");
|
|
26
|
+
}
|
|
27
|
+
const limit = clampLimit(request.limit);
|
|
28
|
+
const filters = request.filters ?? {};
|
|
29
|
+
const offset = request.cursor
|
|
30
|
+
? decodeSearchCursor(request.cursor, { query, filters }).offset
|
|
31
|
+
: 0;
|
|
32
|
+
const taxonomyKeys = request.taxonomy_keys;
|
|
33
|
+
const proximityWeight = request.rrf_weights?.phrase ?? this.proximityWeight;
|
|
34
|
+
const scored = this.chunks
|
|
35
|
+
.filter((chunk) => matchesMetadataFilters(chunk.metadata, filters, taxonomyKeys))
|
|
36
|
+
.map((chunk) => ({ chunk, score: scoreChunk(chunk, query, proximityWeight) }))
|
|
37
|
+
.filter((entry) => entry.score > 0)
|
|
38
|
+
.sort((a, b) => b.score - a.score || a.chunk.chunk_id.localeCompare(b.chunk.chunk_id));
|
|
39
|
+
const paged = scored.slice(offset, offset + limit);
|
|
40
|
+
const hits = paged.map(({ chunk, score }) => toSearchHit(chunk, score, query));
|
|
41
|
+
const nextOffset = offset + paged.length;
|
|
42
|
+
const nextCursor = nextOffset < scored.length
|
|
43
|
+
? encodeSearchCursor({ offset: nextOffset, limit }, { query, filters })
|
|
44
|
+
: null;
|
|
45
|
+
if (hits.length > 0) {
|
|
46
|
+
return {
|
|
47
|
+
hits,
|
|
48
|
+
next_cursor: nextCursor,
|
|
49
|
+
hint: null
|
|
50
|
+
};
|
|
51
|
+
}
|
|
52
|
+
return {
|
|
53
|
+
hits,
|
|
54
|
+
next_cursor: nextCursor,
|
|
55
|
+
hint: buildHint(this.chunks, query, filters)
|
|
56
|
+
};
|
|
57
|
+
}
|
|
58
|
+
async getDoc(request) {
|
|
59
|
+
if (!isChunkIdFormat(request.chunk_id)) {
|
|
60
|
+
throw new Error(`Chunk ID '${request.chunk_id}' has invalid format. Expected {filepath} or {filepath}#{heading-path}.`);
|
|
61
|
+
}
|
|
62
|
+
const context = Math.max(0, Math.min(5, request.context ?? 0));
|
|
63
|
+
const target = this.byId.get(request.chunk_id);
|
|
64
|
+
if (!target) {
|
|
65
|
+
throw new Error(`Chunk ID '${request.chunk_id}' not found. Use search_docs to discover valid chunk IDs.`);
|
|
66
|
+
}
|
|
67
|
+
const fileChunks = this.byFile.get(target.filepath);
|
|
68
|
+
if (!fileChunks) {
|
|
69
|
+
throw new Error(`Internal error: file '${target.filepath}' is missing from index`);
|
|
70
|
+
}
|
|
71
|
+
const indexInFile = fileChunks.findIndex((chunk) => chunk.chunk_id === target.chunk_id);
|
|
72
|
+
const start = Math.max(0, indexInFile - context);
|
|
73
|
+
const end = Math.min(fileChunks.length - 1, indexInFile + context);
|
|
74
|
+
const blocks = [];
|
|
75
|
+
for (let i = start; i <= end; i += 1) {
|
|
76
|
+
const chunk = fileChunks[i];
|
|
77
|
+
if (!chunk) {
|
|
78
|
+
continue;
|
|
79
|
+
}
|
|
80
|
+
const positionLabel = `Chunk ${i + 1} of ${fileChunks.length}`;
|
|
81
|
+
const contextOffset = i - indexInFile;
|
|
82
|
+
const role = contextOffset === 0
|
|
83
|
+
? "Target"
|
|
84
|
+
: `Context: ${contextOffset > 0 ? `+${contextOffset}` : contextOffset}`;
|
|
85
|
+
blocks.push(`--- Chunk: ${chunk.chunk_id} (${positionLabel}) (${role}) ---\n${chunk.content}`);
|
|
86
|
+
}
|
|
87
|
+
return {
|
|
88
|
+
text: blocks.join("\n\n")
|
|
89
|
+
};
|
|
90
|
+
}
|
|
91
|
+
}
|
|
92
|
+
// Backwards-compatible alias while introducing the SearchEngine abstraction.
|
|
93
|
+
export class DocsIndex extends InMemorySearchEngine {
|
|
94
|
+
}
|
|
95
|
+
function scoreChunk(chunk, query, proximityWeight) {
|
|
96
|
+
const queryNormalized = normalizeSearchText(query);
|
|
97
|
+
const tokens = tokenizeSearchText(queryNormalized);
|
|
98
|
+
if (tokens.length === 0) {
|
|
99
|
+
return 0;
|
|
100
|
+
}
|
|
101
|
+
const heading = normalizeSearchText(chunk.heading);
|
|
102
|
+
const content = normalizeSearchText(chunk.content_text);
|
|
103
|
+
let lexical = 0;
|
|
104
|
+
for (const token of tokens) {
|
|
105
|
+
lexical += countToken(content, token) + 3 * countToken(heading, token);
|
|
106
|
+
}
|
|
107
|
+
const phraseMatch = content.includes(queryNormalized) || heading.includes(queryNormalized);
|
|
108
|
+
const proximityBoost = phraseMatch ? proximityWeight : 0;
|
|
109
|
+
return lexical + proximityBoost;
|
|
110
|
+
}
|
|
111
|
+
function toSearchHit(chunk, score, query) {
|
|
112
|
+
return {
|
|
113
|
+
chunk_id: chunk.chunk_id,
|
|
114
|
+
heading: chunk.heading,
|
|
115
|
+
breadcrumb: chunk.breadcrumb,
|
|
116
|
+
snippet: makeSnippet(chunk.content_text, query),
|
|
117
|
+
filepath: chunk.filepath,
|
|
118
|
+
metadata: chunk.metadata,
|
|
119
|
+
score: Number(score.toFixed(6))
|
|
120
|
+
};
|
|
121
|
+
}
|
|
122
|
+
function buildHint(chunks, query, filters) {
|
|
123
|
+
const queryMatches = chunks.filter((chunk) => scoreChunk(chunk, query, 0) > 0);
|
|
124
|
+
if (queryMatches.length === 0) {
|
|
125
|
+
return {
|
|
126
|
+
message: "0 results found. No matches were found for this query in the indexed corpus.",
|
|
127
|
+
suggested_filters: {}
|
|
128
|
+
};
|
|
129
|
+
}
|
|
130
|
+
const suggestions = {};
|
|
131
|
+
for (const [key, activeValue] of Object.entries(filters)) {
|
|
132
|
+
const values = new Set();
|
|
133
|
+
for (const chunk of queryMatches) {
|
|
134
|
+
const candidate = chunk.metadata[key];
|
|
135
|
+
if (!candidate || candidate === activeValue) {
|
|
136
|
+
continue;
|
|
137
|
+
}
|
|
138
|
+
values.add(candidate);
|
|
139
|
+
}
|
|
140
|
+
if (values.size > 0) {
|
|
141
|
+
suggestions[key] = [...values].sort((a, b) => a.localeCompare(b));
|
|
142
|
+
}
|
|
143
|
+
}
|
|
144
|
+
const filterSummary = Object.entries(filters)
|
|
145
|
+
.map(([key, value]) => `${key}='${value}'`)
|
|
146
|
+
.join(", ");
|
|
147
|
+
return {
|
|
148
|
+
message: filterSummary
|
|
149
|
+
? `0 results found for query '${query}' with filters ${filterSummary}.`
|
|
150
|
+
: `0 results found for query '${query}'.`,
|
|
151
|
+
suggested_filters: suggestions
|
|
152
|
+
};
|
|
153
|
+
}
|
|
154
|
+
function countToken(value, token) {
|
|
155
|
+
if (!token) {
|
|
156
|
+
return 0;
|
|
157
|
+
}
|
|
158
|
+
let count = 0;
|
|
159
|
+
let index = value.indexOf(token);
|
|
160
|
+
while (index >= 0) {
|
|
161
|
+
count += 1;
|
|
162
|
+
index = value.indexOf(token, index + token.length);
|
|
163
|
+
}
|
|
164
|
+
return count;
|
|
165
|
+
}
|
|
166
|
+
//# sourceMappingURL=search.js.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"search.js","sourceRoot":"","sources":["../src/search.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,kBAAkB,EAAE,kBAAkB,EAAE,MAAM,aAAa,CAAC;AACrE,OAAO,EACL,UAAU,EACV,eAAe,EACf,WAAW,EACX,sBAAsB,EACtB,mBAAmB,EACnB,kBAAkB,EACnB,MAAM,oBAAoB,CAAC;AAa5B,MAAM,OAAO,oBAAoB;IACd,MAAM,CAAU;IAChB,IAAI,CAAqB;IACzB,MAAM,CAAuB;IAC7B,eAAe,CAAS;IAEzC,YAAY,MAAe,EAAE,UAA4B,EAAE;QACzD,IAAI,CAAC,MAAM,GAAG,CAAC,GAAG,MAAM,CAAC,CAAC;QAC1B,IAAI,CAAC,IAAI,GAAG,IAAI,GAAG,CAAC,MAAM,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC,KAAK,CAAC,QAAQ,EAAE,KAAK,CAAC,CAAC,CAAC,CAAC;QACpE,IAAI,CAAC,MAAM,GAAG,IAAI,GAAG,EAAmB,CAAC;QACzC,IAAI,CAAC,eAAe,GAAG,OAAO,CAAC,eAAe,IAAI,IAAI,CAAC;QAEvD,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;YAC3B,MAAM,IAAI,GAAG,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,KAAK,CAAC,QAAQ,CAAC,IAAI,EAAE,CAAC;YACnD,IAAI,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;YACjB,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,KAAK,CAAC,QAAQ,EAAE,IAAI,CAAC,CAAC;QACxC,CAAC;QAED,KAAK,MAAM,IAAI,IAAI,IAAI,CAAC,MAAM,CAAC,MAAM,EAAE,EAAE,CAAC;YACxC,IAAI,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,WAAW,GAAG,CAAC,CAAC,WAAW,CAAC,CAAC;QACrD,CAAC;IACH,CAAC;IAED,KAAK,CAAC,MAAM,CAAC,OAAsB;QACjC,MAAM,KAAK,GAAG,OAAO,CAAC,KAAK,CAAC,IAAI,EAAE,CAAC;QACnC,IAAI,CAAC,KAAK,EAAE,CAAC;YACX,MAAM,IAAI,KAAK,CAAC,mBAAmB,CAAC,CAAC;QACvC,CAAC;QAED,MAAM,KAAK,GAAG,UAAU,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC;QACxC,MAAM,OAAO,GAAG,OAAO,CAAC,OAAO,IAAI,EAAE,CAAC;QACtC,MAAM,MAAM,GAAG,OAAO,CAAC,MAAM;YAC3B,CAAC,CAAC,kBAAkB,CAAC,OAAO,CAAC,MAAM,EAAE,EAAE,KAAK,EAAE,OAAO,EAAE,CAAC,CAAC,MAAM;YAC/D,CAAC,CAAC,CAAC,CAAC;QAEN,MAAM,YAAY,GAAG,OAAO,CAAC,aAAa,CAAC;QAC3C,MAAM,eAAe,GAAG,OAAO,CAAC,WAAW,EAAE,MAAM,IAAI,IAAI,CAAC,eAAe,CAAC;QAE5E,MAAM,MAAM,GAAG,IAAI,CAAC,MAAM;aACvB,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,sBAAsB,CAAC,KAAK,CAAC,QAAQ,EAAE,OAAO,EAAE,YAAY,CAAC,CAAC;aAChF,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,CAAC,EAAE,KAAK,EAAE,KAAK,EAAE,UAAU,CAAC,KAAK,EAAE,KAAK,EAAE,eAAe,CAAC,EAAE,CAAC,CAAC;aAC7E,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,KAAK,GAAG,CAAC,CAAC;aAClC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,KAAK,GAAG,CAAC,CAAC,KAAK,IAAI,CAAC,CAAC,KAAK,CAAC,QAAQ,CAAC,aAAa,CAAC,CAAC,CAAC,KAAK,CAAC,QAAQ,CAAC,CAAC,CAAC;QAEzF,MAAM,KAAK,GAAG,MAAM,CAAC,KAAK,CAAC,MAAM,EAAE,MAAM,GAAG,KAAK,CAAC,CAAC;QACnD,MAAM,IAAI,GAAG,KAAK,CAAC,GAAG,CAAC,CAAC,EAAE,KAAK,EAAE,KAAK,EAAE,EAAE,EAAE,CAAC,WAAW,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAAC,CAAC;QAE/E,MAAM,UAAU,GAAG,MAAM,GAAG,KAAK,CAAC,MAAM,CAAC;QACzC,MAAM,UAAU,GAAG,UAAU,GAAG,MAAM,CAAC,MAAM;YAC3C,CAAC,CAAC,kBAAkB,CAAC,EAAE,MAAM,EAAE,UAAU,EAAE,KAAK,EAAE,EAAE,EAAE,KAAK,EAAE,OAAO,EAAE,CAAC;YACvE,CAAC,CAAC,IAAI,CAAC;QAET,IAAI,IAAI,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACpB,OAAO;gBACL,IAAI;gBACJ,WAAW,EAAE,UAAU;gBACvB,IAAI,EAAE,IAAI;aACX,CAAC;QACJ,CAAC;QAED,OAAO;YACL,IAAI;YACJ,WAAW,EAAE,UAAU;YACvB,IAAI,EAAE,SAAS,CAAC,IAAI,CAAC,MAAM,EAAE,KAAK,EAAE,OAAO,CAAC;SAC7C,CAAC;IACJ,CAAC;IAED,KAAK,CAAC,MAAM,CAAC,OAAsB;QACjC,IAAI,CAAC,eAAe,CAAC,OAAO,CAAC,QAAQ,CAAC,EAAE,CAAC;YACvC,MAAM,IAAI,KAAK,CACb,aAAa,OAAO,CAAC,QAAQ,yEAAyE,CACvG,CAAC;QACJ,CAAC;QAED,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,OAAO,CAAC,OAAO,IAAI,CAAC,CAAC,CAAC,CAAC;QAC/D,MAAM,MAAM,GAAG,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAC;QAC/C,IAAI,CAAC,MAAM,EAAE,CAAC;YACZ,MAAM,IAAI,KAAK,CACb,aAAa,OAAO,CAAC,QAAQ,2DAA2D,CACzF,CAAC;QACJ,CAAC;QAED,MAAM,UAAU,GAAG,IAAI,CAAC,MAAM,CAAC,GAAG,CAAC,MAAM,CAAC,QAAQ,CAAC,CAAC;QACpD,IAAI,CAAC,UAAU,EAAE,CAAC;YAChB,MAAM,IAAI,KAAK,CAAC,yBAAyB,MAAM,CAAC,QAAQ,yBAAyB,CAAC,CAAC;QACrF,CAAC;QAED,MAAM,WAAW,GAAG,UAAU,CAAC,SAAS,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,QAAQ,KAAK,MAAM,CAAC,QAAQ,CAAC,CAAC;QACxF,MAAM,KAAK,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,WAAW,GAAG,OAAO,CAAC,CAAC;QACjD,MAAM,GAAG,GAAG,IAAI,CAAC,GAAG,CAAC,UAAU,CAAC,MAAM,GAAG,CAAC,EAAE,WAAW,GAAG,OAAO,CAAC,CAAC;QAEnE,MAAM,MAAM,GAAa,EAAE,CAAC;QAC5B,KAAK,IAAI,CAAC,GAAG,KAAK,EAAE,CAAC,IAAI,GAAG,EAAE,CAAC,IAAI,CAAC,EAAE,CAAC;YACrC,MAAM,KAAK,GAAG,UAAU,CAAC,CAAC,CAAC,CAAC;YAC5B,IAAI,CAAC,KAAK,EAAE,CAAC;gBACX,SAAS;YACX,CAAC;YACD,MAAM,aAAa,GAAG,SAAS,CAAC,GAAG,CAAC,OAAO,UAAU,CAAC,MAAM,EAAE,CAAC;YAC/D,MAAM,aAAa,GAAG,CAAC,GAAG,WAAW,CAAC;YACtC,MAAM,IAAI,GACR,aAAa,KAAK,CAAC;gBACjB,CAAC,CAAC,QAAQ;gBACV,CAAC,CAAC,YAAY,aAAa,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,aAAa,EAAE,CAAC,CAAC,CAAC,aAAa,EAAE,CAAC;YAE5E,MAAM,CAAC,IAAI,CACT,cAAc,KAAK,CAAC,QAAQ,KAAK,aAAa,MAAM,IAAI,UAAU,KAAK,CAAC,OAAO,EAAE,CAClF,CAAC;QACJ,CAAC;QAED,OAAO;YACL,IAAI,EAAE,MAAM,CAAC,IAAI,CAAC,MAAM,CAAC;SAC1B,CAAC;IACJ,CAAC;CACF;AAED,6EAA6E;AAC7E,MAAM,OAAO,SAAU,SAAQ,oBAAoB;CAAG;AAEtD,SAAS,UAAU,CAAC,KAAY,EAAE,KAAa,EAAE,eAAuB;IACtE,MAAM,eAAe,GAAG,mBAAmB,CAAC,KAAK,CAAC,CAAC;IACnD,MAAM,MAAM,GAAG,kBAAkB,CAAC,eAAe,CAAC,CAAC;IACnD,IAAI,MAAM,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACxB,OAAO,CAAC,CAAC;IACX,CAAC;IAED,MAAM,OAAO,GAAG,mBAAmB,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC;IACnD,MAAM,OAAO,GAAG,mBAAmB,CAAC,KAAK,CAAC,YAAY,CAAC,CAAC;IAExD,IAAI,OAAO,GAAG,CAAC,CAAC;IAChB,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;QAC3B,OAAO,IAAI,UAAU,CAAC,OAAO,EAAE,KAAK,CAAC,GAAG,CAAC,GAAG,UAAU,CAAC,OAAO,EAAE,KAAK,CAAC,CAAC;IACzE,CAAC;IAED,MAAM,WAAW,GAAG,OAAO,CAAC,QAAQ,CAAC,eAAe,CAAC,IAAI,OAAO,CAAC,QAAQ,CAAC,eAAe,CAAC,CAAC;IAC3F,MAAM,cAAc,GAAG,WAAW,CAAC,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC,CAAC;IAEzD,OAAO,OAAO,GAAG,cAAc,CAAC;AAClC,CAAC;AAED,SAAS,WAAW,CAAC,KAAY,EAAE,KAAa,EAAE,KAAa;IAC7D,OAAO;QACL,QAAQ,EAAE,KAAK,CAAC,QAAQ;QACxB,OAAO,EAAE,KAAK,CAAC,OAAO;QACtB,UAAU,EAAE,KAAK,CAAC,UAAU;QAC5B,OAAO,EAAE,WAAW,CAAC,KAAK,CAAC,YAAY,EAAE,KAAK,CAAC;QAC/C,QAAQ,EAAE,KAAK,CAAC,QAAQ;QACxB,QAAQ,EAAE,KAAK,CAAC,QAAQ;QACxB,KAAK,EAAE,MAAM,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,CAAC;KAChC,CAAC;AACJ,CAAC;AAED,SAAS,SAAS,CAChB,MAAe,EACf,KAAa,EACb,OAA+B;IAE/B,MAAM,YAAY,GAAG,MAAM,CAAC,MAAM,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,UAAU,CAAC,KAAK,EAAE,KAAK,EAAE,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC;IAC/E,IAAI,YAAY,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QAC9B,OAAO;YACL,OAAO,EAAE,8EAA8E;YACvF,iBAAiB,EAAE,EAAE;SACtB,CAAC;IACJ,CAAC;IAED,MAAM,WAAW,GAA6B,EAAE,CAAC;IACjD,KAAK,MAAM,CAAC,GAAG,EAAE,WAAW,CAAC,IAAI,MAAM,CAAC,OAAO,CAAC,OAAO,CAAC,EAAE,CAAC;QACzD,MAAM,MAAM,GAAG,IAAI,GAAG,EAAU,CAAC;QACjC,KAAK,MAAM,KAAK,IAAI,YAAY,EAAE,CAAC;YACjC,MAAM,SAAS,GAAG,KAAK,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC;YACtC,IAAI,CAAC,SAAS,IAAI,SAAS,KAAK,WAAW,EAAE,CAAC;gBAC5C,SAAS;YACX,CAAC;YACD,MAAM,CAAC,GAAG,CAAC,SAAS,CAAC,CAAC;QACxB,CAAC;QAED,IAAI,MAAM,CAAC,IAAI,GAAG,CAAC,EAAE,CAAC;YACpB,WAAW,CAAC,GAAG,CAAC,GAAG,CAAC,GAAG,MAAM,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,aAAa,CAAC,CAAC,CAAC,CAAC,CAAC;QACpE,CAAC;IACH,CAAC;IAED,MAAM,aAAa,GAAG,MAAM,CAAC,OAAO,CAAC,OAAO,CAAC;SAC1C,GAAG,CAAC,CAAC,CAAC,GAAG,EAAE,KAAK,CAAC,EAAE,EAAE,CAAC,GAAG,GAAG,KAAK,KAAK,GAAG,CAAC;SAC1C,IAAI,CAAC,IAAI,CAAC,CAAC;IAEd,OAAO;QACL,OAAO,EAAE,aAAa;YACpB,CAAC,CAAC,8BAA8B,KAAK,kBAAkB,aAAa,GAAG;YACvE,CAAC,CAAC,8BAA8B,KAAK,IAAI;QAC3C,iBAAiB,EAAE,WAAW;KAC/B,CAAC;AACJ,CAAC;AAED,SAAS,UAAU,CAAC,KAAa,EAAE,KAAa;IAC9C,IAAI,CAAC,KAAK,EAAE,CAAC;QACX,OAAO,CAAC,CAAC;IACX,CAAC;IAED,IAAI,KAAK,GAAG,CAAC,CAAC;IACd,IAAI,KAAK,GAAG,KAAK,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC;IACjC,OAAO,KAAK,IAAI,CAAC,EAAE,CAAC;QAClB,KAAK,IAAI,CAAC,CAAC;QACX,KAAK,GAAG,KAAK,CAAC,OAAO,CAAC,KAAK,EAAE,KAAK,GAAG,KAAK,CAAC,MAAM,CAAC,CAAC;IACrD,CAAC;IACD,OAAO,KAAK,CAAC;AACf,CAAC"}
|
package/dist/types.d.ts
ADDED
|
@@ -0,0 +1,125 @@
|
|
|
1
|
+
export interface Chunk {
|
|
2
|
+
chunk_id: string;
|
|
3
|
+
filepath: string;
|
|
4
|
+
heading: string;
|
|
5
|
+
heading_level: number;
|
|
6
|
+
content: string;
|
|
7
|
+
content_text: string;
|
|
8
|
+
breadcrumb: string;
|
|
9
|
+
chunk_index: number;
|
|
10
|
+
metadata: Record<string, string>;
|
|
11
|
+
}
|
|
12
|
+
export interface ChunkingStrategy {
|
|
13
|
+
chunk_by: "h1" | "h2" | "h3" | "file";
|
|
14
|
+
max_chunk_size?: number;
|
|
15
|
+
min_chunk_size?: number;
|
|
16
|
+
}
|
|
17
|
+
export interface ManifestOverride {
|
|
18
|
+
pattern: string;
|
|
19
|
+
strategy?: ChunkingStrategy;
|
|
20
|
+
metadata?: Record<string, string>;
|
|
21
|
+
}
|
|
22
|
+
export interface Manifest {
|
|
23
|
+
version: string;
|
|
24
|
+
strategy?: ChunkingStrategy;
|
|
25
|
+
metadata?: Record<string, string>;
|
|
26
|
+
overrides?: ManifestOverride[];
|
|
27
|
+
}
|
|
28
|
+
export interface RrfWeights {
|
|
29
|
+
vector?: number;
|
|
30
|
+
match?: number;
|
|
31
|
+
phrase?: number;
|
|
32
|
+
}
|
|
33
|
+
export interface SearchRequest {
|
|
34
|
+
query: string;
|
|
35
|
+
limit: number;
|
|
36
|
+
cursor?: string;
|
|
37
|
+
filters: Record<string, string>;
|
|
38
|
+
/** Available taxonomy keys from metadata, used for conditional auto-include */
|
|
39
|
+
taxonomy_keys?: string[];
|
|
40
|
+
/** Optional RRF weight overrides for this request */
|
|
41
|
+
rrf_weights?: RrfWeights;
|
|
42
|
+
}
|
|
43
|
+
export interface SearchHit {
|
|
44
|
+
chunk_id: string;
|
|
45
|
+
heading: string;
|
|
46
|
+
breadcrumb: string;
|
|
47
|
+
snippet: string;
|
|
48
|
+
filepath: string;
|
|
49
|
+
metadata: Record<string, string>;
|
|
50
|
+
score: number;
|
|
51
|
+
}
|
|
52
|
+
export interface SearchHint {
|
|
53
|
+
message: string;
|
|
54
|
+
suggested_filters: Record<string, string[]>;
|
|
55
|
+
}
|
|
56
|
+
export interface SearchResult {
|
|
57
|
+
hits: SearchHit[];
|
|
58
|
+
next_cursor: string | null;
|
|
59
|
+
hint: SearchHint | null;
|
|
60
|
+
}
|
|
61
|
+
export interface SearchEngine {
|
|
62
|
+
search(request: SearchRequest): Promise<SearchResult>;
|
|
63
|
+
getDoc(request: GetDocRequest): Promise<GetDocResult>;
|
|
64
|
+
}
|
|
65
|
+
export interface TaxonomyField {
|
|
66
|
+
description?: string;
|
|
67
|
+
values: string[];
|
|
68
|
+
}
|
|
69
|
+
export interface MetadataStats {
|
|
70
|
+
total_chunks: number;
|
|
71
|
+
total_files: number;
|
|
72
|
+
indexed_at: string;
|
|
73
|
+
source_commit?: string | null;
|
|
74
|
+
}
|
|
75
|
+
export interface EmbeddingMetadata {
|
|
76
|
+
provider: string;
|
|
77
|
+
model: string;
|
|
78
|
+
dimensions: number;
|
|
79
|
+
}
|
|
80
|
+
export interface CorpusMetadata {
|
|
81
|
+
metadata_version: string;
|
|
82
|
+
corpus_description: string;
|
|
83
|
+
taxonomy: Record<string, TaxonomyField>;
|
|
84
|
+
stats: MetadataStats;
|
|
85
|
+
embedding: EmbeddingMetadata | null;
|
|
86
|
+
}
|
|
87
|
+
export interface GetDocRequest {
|
|
88
|
+
chunk_id: string;
|
|
89
|
+
context?: number;
|
|
90
|
+
}
|
|
91
|
+
export interface GetDocResult {
|
|
92
|
+
text: string;
|
|
93
|
+
}
|
|
94
|
+
export interface BuildChunksInput {
|
|
95
|
+
filepath: string;
|
|
96
|
+
markdown: string;
|
|
97
|
+
strategy: ChunkingStrategy;
|
|
98
|
+
metadata?: Record<string, string>;
|
|
99
|
+
}
|
|
100
|
+
export interface ResolvedFileConfig {
|
|
101
|
+
strategy: ChunkingStrategy;
|
|
102
|
+
metadata: Record<string, string>;
|
|
103
|
+
}
|
|
104
|
+
export interface EmbeddingConfig {
|
|
105
|
+
provider: string;
|
|
106
|
+
model: string;
|
|
107
|
+
dimensions: number;
|
|
108
|
+
configFingerprint: string;
|
|
109
|
+
}
|
|
110
|
+
export interface EmbeddingProvider {
|
|
111
|
+
readonly name: string;
|
|
112
|
+
readonly model: string;
|
|
113
|
+
readonly dimensions: number;
|
|
114
|
+
readonly configFingerprint: string;
|
|
115
|
+
embed(texts: string[]): Promise<number[][]>;
|
|
116
|
+
}
|
|
117
|
+
export interface DocsIndexOptions {
|
|
118
|
+
proximityWeight?: number;
|
|
119
|
+
phraseSlop?: number;
|
|
120
|
+
}
|
|
121
|
+
export interface CursorPayload {
|
|
122
|
+
offset: number;
|
|
123
|
+
limit: number;
|
|
124
|
+
}
|
|
125
|
+
//# sourceMappingURL=types.d.ts.map
|
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"types.d.ts","sourceRoot":"","sources":["../src/types.ts"],"names":[],"mappings":"AAAA,MAAM,WAAW,KAAK;IACpB,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,MAAM,CAAC;IACjB,OAAO,EAAE,MAAM,CAAC;IAChB,aAAa,EAAE,MAAM,CAAC;IACtB,OAAO,EAAE,MAAM,CAAC;IAChB,YAAY,EAAE,MAAM,CAAC;IACrB,UAAU,EAAE,MAAM,CAAC;IACnB,WAAW,EAAE,MAAM,CAAC;IACpB,QAAQ,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CAClC;AAED,MAAM,WAAW,gBAAgB;IAC/B,QAAQ,EAAE,IAAI,GAAG,IAAI,GAAG,IAAI,GAAG,MAAM,CAAC;IACtC,cAAc,CAAC,EAAE,MAAM,CAAC;IACxB,cAAc,CAAC,EAAE,MAAM,CAAC;CACzB;AAED,MAAM,WAAW,gBAAgB;IAC/B,OAAO,EAAE,MAAM,CAAC;IAChB,QAAQ,CAAC,EAAE,gBAAgB,CAAC;IAC5B,QAAQ,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CACnC;AAED,MAAM,WAAW,QAAQ;IACvB,OAAO,EAAE,MAAM,CAAC;IAChB,QAAQ,CAAC,EAAE,gBAAgB,CAAC;IAC5B,QAAQ,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IAClC,SAAS,CAAC,EAAE,gBAAgB,EAAE,CAAC;CAChC;AAED,MAAM,WAAW,UAAU;IACzB,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,KAAK,CAAC,EAAE,MAAM,CAAC;IACf,MAAM,CAAC,EAAE,MAAM,CAAC;CACjB;AAED,MAAM,WAAW,aAAa;IAC5B,KAAK,EAAE,MAAM,CAAC;IACd,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,CAAC,EAAE,MAAM,CAAC;IAChB,OAAO,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IAChC,+EAA+E;IAC/E,aAAa,CAAC,EAAE,MAAM,EAAE,CAAC;IACzB,qDAAqD;IACrD,WAAW,CAAC,EAAE,UAAU,CAAC;CAC1B;AAED,MAAM,WAAW,SAAS;IACxB,QAAQ,EAAE,MAAM,CAAC;IACjB,OAAO,EAAE,MAAM,CAAC;IAChB,UAAU,EAAE,MAAM,CAAC;IACnB,OAAO,EAAE,MAAM,CAAC;IAChB,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;IACjC,KAAK,EAAE,MAAM,CAAC;CACf;AAED,MAAM,WAAW,UAAU;IACzB,OAAO,EAAE,MAAM,CAAC;IAChB,iBAAiB,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,EAAE,CAAC,CAAC;CAC7C;AAED,MAAM,WAAW,YAAY;IAC3B,IAAI,EAAE,SAAS,EAAE,CAAC;IAClB,WAAW,EAAE,MAAM,GAAG,IAAI,CAAC;IAC3B,IAAI,EAAE,UAAU,GAAG,IAAI,CAAC;CACzB;AAED,MAAM,WAAW,YAAY;IAC3B,MAAM,CAAC,OAAO,EAAE,aAAa,GAAG,OAAO,CAAC,YAAY,CAAC,CAAC;IACtD,MAAM,CAAC,OAAO,EAAE,aAAa,GAAG,OAAO,CAAC,YAAY,CAAC,CAAC;CACvD;AAED,MAAM,WAAW,aAAa;IAC5B,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,MAAM,EAAE,MAAM,EAAE,CAAC;CAClB;AAED,MAAM,WAAW,aAAa;IAC5B,YAAY,EAAE,MAAM,CAAC;IACrB,WAAW,EAAE,MAAM,CAAC;IACpB,UAAU,EAAE,MAAM,CAAC;IACnB,aAAa,CAAC,EAAE,MAAM,GAAG,IAAI,CAAC;CAC/B;AAED,MAAM,WAAW,iBAAiB;IAChC,QAAQ,EAAE,MAAM,CAAC;IACjB,KAAK,EAAE,MAAM,CAAC;IACd,UAAU,EAAE,MAAM,CAAC;CACpB;AAED,MAAM,WAAW,cAAc;IAC7B,gBAAgB,EAAE,MAAM,CAAC;IACzB,kBAAkB,EAAE,MAAM,CAAC;IAC3B,QAAQ,EAAE,MAAM,CAAC,MAAM,EAAE,aAAa,CAAC,CAAC;IACxC,KAAK,EAAE,aAAa,CAAC;IACrB,SAAS,EAAE,iBAAiB,GAAG,IAAI,CAAC;CACrC;AAED,MAAM,WAAW,aAAa;IAC5B,QAAQ,EAAE,MAAM,CAAC;IACjB,OAAO,CAAC,EAAE,MAAM,CAAC;CAClB;AAED,MAAM,WAAW,YAAY;IAC3B,IAAI,EAAE,MAAM,CAAC;CACd;AAED,MAAM,WAAW,gBAAgB;IAC/B,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,MAAM,CAAC;IACjB,QAAQ,EAAE,gBAAgB,CAAC;IAC3B,QAAQ,CAAC,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CACnC;AAED,MAAM,WAAW,kBAAkB;IACjC,QAAQ,EAAE,gBAAgB,CAAC;IAC3B,QAAQ,EAAE,MAAM,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CAClC;AAED,MAAM,WAAW,eAAe;IAC9B,QAAQ,EAAE,MAAM,CAAC;IACjB,KAAK,EAAE,MAAM,CAAC;IACd,UAAU,EAAE,MAAM,CAAC;IACnB,iBAAiB,EAAE,MAAM,CAAC;CAC3B;AAED,MAAM,WAAW,iBAAiB;IAChC,QAAQ,CAAC,IAAI,EAAE,MAAM,CAAC;IACtB,QAAQ,CAAC,KAAK,EAAE,MAAM,CAAC;IACvB,QAAQ,CAAC,UAAU,EAAE,MAAM,CAAC;IAC5B,QAAQ,CAAC,iBAAiB,EAAE,MAAM,CAAC;IACnC,KAAK,CAAC,KAAK,EAAE,MAAM,EAAE,GAAG,OAAO,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC;CAC7C;AAED,MAAM,WAAW,gBAAgB;IAC/B,eAAe,CAAC,EAAE,MAAM,CAAC;IACzB,UAAU,CAAC,EAAE,MAAM,CAAC;CACrB;AAED,MAAM,WAAW,aAAa;IAC5B,MAAM,EAAE,MAAM,CAAC;IACf,KAAK,EAAE,MAAM,CAAC;CACf"}
|
package/dist/types.js
ADDED
|
@@ -0,0 +1 @@
|
|
|
1
|
+
{"version":3,"file":"types.js","sourceRoot":"","sources":["../src/types.ts"],"names":[],"mappings":""}
|
package/package.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
1
|
+
{
|
|
2
|
+
"name": "@speakeasy-api/docs-mcp-core",
|
|
3
|
+
"version": "0.1.0",
|
|
4
|
+
"description": "Core retrieval and indexing primitives for docs-mcp: markdown chunking, hybrid search, and LanceDB integration",
|
|
5
|
+
"license": "AGPL-3.0-only",
|
|
6
|
+
"type": "module",
|
|
7
|
+
"author": "Speakeasy <info@speakeasy.com> (https://speakeasy.com)",
|
|
8
|
+
"repository": {
|
|
9
|
+
"type": "git",
|
|
10
|
+
"url": "https://github.com/speakeasy-api/docs-mcp.git",
|
|
11
|
+
"directory": "packages/core"
|
|
12
|
+
},
|
|
13
|
+
"homepage": "https://github.com/speakeasy-api/docs-mcp",
|
|
14
|
+
"bugs": {
|
|
15
|
+
"url": "https://github.com/speakeasy-api/docs-mcp/issues"
|
|
16
|
+
},
|
|
17
|
+
"files": [
|
|
18
|
+
"dist"
|
|
19
|
+
],
|
|
20
|
+
"main": "dist/index.js",
|
|
21
|
+
"types": "dist/index.d.ts",
|
|
22
|
+
"exports": {
|
|
23
|
+
".": {
|
|
24
|
+
"types": "./dist/index.d.ts",
|
|
25
|
+
"import": "./dist/index.js"
|
|
26
|
+
}
|
|
27
|
+
},
|
|
28
|
+
"engines": {
|
|
29
|
+
"node": ">=22.0.0"
|
|
30
|
+
},
|
|
31
|
+
"dependencies": {
|
|
32
|
+
"@lancedb/lancedb": "^0.26.2",
|
|
33
|
+
"apache-arrow": "^18.1.0",
|
|
34
|
+
"gray-matter": "^4.0.3",
|
|
35
|
+
"mdast-util-to-string": "^4.0.0",
|
|
36
|
+
"picomatch": "^4.0.2",
|
|
37
|
+
"remark-frontmatter": "^5.0.0",
|
|
38
|
+
"remark-gfm": "^4.0.0",
|
|
39
|
+
"remark-parse": "^11.0.0",
|
|
40
|
+
"semver": "^7.7.1",
|
|
41
|
+
"unified": "^11.0.5"
|
|
42
|
+
},
|
|
43
|
+
"scripts": {
|
|
44
|
+
"build": "tsc -p tsconfig.json",
|
|
45
|
+
"typecheck": "tsc -p tsconfig.json --noEmit",
|
|
46
|
+
"test": "vitest run",
|
|
47
|
+
"lint": "eslint src test"
|
|
48
|
+
}
|
|
49
|
+
}
|