@cogitator-ai/rag 0.1.10 → 0.3.0
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/README.md +66 -31
- package/dist/chunkers/create-chunker.js.map +1 -1
- package/dist/chunkers/index.d.ts.map +1 -1
- package/dist/chunkers/index.js.map +1 -1
- package/dist/chunkers/recursive-chunker.d.ts +1 -2
- package/dist/chunkers/recursive-chunker.d.ts.map +1 -1
- package/dist/chunkers/recursive-chunker.js +68 -58
- package/dist/chunkers/recursive-chunker.js.map +1 -1
- package/dist/chunkers/semantic-chunker.d.ts.map +1 -1
- package/dist/chunkers/semantic-chunker.js +5 -2
- package/dist/chunkers/semantic-chunker.js.map +1 -1
- package/dist/index.d.ts +2 -2
- package/dist/index.d.ts.map +1 -1
- package/dist/index.js +4 -2
- package/dist/index.js.map +1 -1
- package/dist/loaders/csv-loader.d.ts.map +1 -1
- package/dist/loaders/csv-loader.js +9 -4
- package/dist/loaders/csv-loader.js.map +1 -1
- package/dist/loaders/html-loader.d.ts +4 -0
- package/dist/loaders/html-loader.d.ts.map +1 -1
- package/dist/loaders/html-loader.js +26 -2
- package/dist/loaders/html-loader.js.map +1 -1
- package/dist/loaders/index.d.ts.map +1 -1
- package/dist/loaders/index.js.map +1 -1
- package/dist/loaders/json-loader.js +1 -1
- package/dist/loaders/json-loader.js.map +1 -1
- package/dist/loaders/markdown-loader.d.ts.map +1 -1
- package/dist/loaders/markdown-loader.js +7 -3
- package/dist/loaders/markdown-loader.js.map +1 -1
- package/dist/loaders/pdf-loader.d.ts +3 -3
- package/dist/loaders/pdf-loader.d.ts.map +1 -1
- package/dist/loaders/pdf-loader.js +45 -42
- package/dist/loaders/pdf-loader.js.map +1 -1
- package/dist/loaders/text-loader.js +1 -1
- package/dist/loaders/text-loader.js.map +1 -1
- package/dist/loaders/web-loader.d.ts +29 -1
- package/dist/loaders/web-loader.d.ts.map +1 -1
- package/dist/loaders/web-loader.js +227 -92
- package/dist/loaders/web-loader.js.map +1 -1
- package/dist/rag-builder.d.ts +10 -1
- package/dist/rag-builder.d.ts.map +1 -1
- package/dist/rag-builder.js +70 -7
- package/dist/rag-builder.js.map +1 -1
- package/dist/rag-pipeline.d.ts +1 -0
- package/dist/rag-pipeline.d.ts.map +1 -1
- package/dist/rag-pipeline.js +55 -19
- package/dist/rag-pipeline.js.map +1 -1
- package/dist/rerankers/index.d.ts.map +1 -1
- package/dist/rerankers/index.js.map +1 -1
- package/dist/retrievers/chunk-indexer.d.ts +19 -0
- package/dist/retrievers/chunk-indexer.d.ts.map +1 -0
- package/dist/retrievers/chunk-indexer.js +8 -0
- package/dist/retrievers/chunk-indexer.js.map +1 -0
- package/dist/retrievers/create-retriever.d.ts +1 -0
- package/dist/retrievers/create-retriever.d.ts.map +1 -1
- package/dist/retrievers/create-retriever.js +1 -0
- package/dist/retrievers/create-retriever.js.map +1 -1
- package/dist/retrievers/hybrid-retriever.d.ts +8 -1
- package/dist/retrievers/hybrid-retriever.d.ts.map +1 -1
- package/dist/retrievers/hybrid-retriever.js +18 -5
- package/dist/retrievers/hybrid-retriever.js.map +1 -1
- package/dist/retrievers/index.d.ts +2 -0
- package/dist/retrievers/index.d.ts.map +1 -1
- package/dist/retrievers/index.js +1 -0
- package/dist/retrievers/index.js.map +1 -1
- package/dist/retrievers/mmr-retriever.d.ts +5 -0
- package/dist/retrievers/mmr-retriever.d.ts.map +1 -1
- package/dist/retrievers/mmr-retriever.js +24 -2
- package/dist/retrievers/mmr-retriever.js.map +1 -1
- package/dist/retrievers/multi-query-retriever.d.ts +12 -1
- package/dist/retrievers/multi-query-retriever.d.ts.map +1 -1
- package/dist/retrievers/multi-query-retriever.js +46 -7
- package/dist/retrievers/multi-query-retriever.js.map +1 -1
- package/dist/retrievers/similarity-retriever.d.ts.map +1 -1
- package/dist/retrievers/similarity-retriever.js +2 -1
- package/dist/retrievers/similarity-retriever.js.map +1 -1
- package/dist/tools.d.ts +11 -2
- package/dist/tools.d.ts.map +1 -1
- package/dist/tools.js +39 -7
- package/dist/tools.js.map +1 -1
- package/package.json +15 -14
package/dist/tools.d.ts.map
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"tools.d.ts","sourceRoot":"","sources":["../src/tools.ts"],"names":[],"mappings":"
|
|
1
|
+
{"version":3,"file":"tools.d.ts","sourceRoot":"","sources":["../src/tools.ts"],"names":[],"mappings":"AAEA,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AACxB,OAAO,KAAK,EAAE,WAAW,EAAE,MAAM,mBAAmB,CAAC;AAErD,MAAM,WAAW,OAAO,CAAC,OAAO,GAAG,OAAO;IACxC,IAAI,EAAE,MAAM,CAAC;IACb,WAAW,EAAE,MAAM,CAAC;IACpB,UAAU,EAAE,CAAC,CAAC,OAAO,CAAC,OAAO,CAAC,CAAC;IAC/B,OAAO,EAAE,CAAC,MAAM,EAAE,OAAO,KAAK,OAAO,CAAC,OAAO,CAAC,CAAC;CAChD;AAED,QAAA,MAAM,kBAAkB;;;;iBAItB,CAAC;AAEH,KAAK,YAAY,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,kBAAkB,CAAC,CAAC;AAEvD,QAAA,MAAM,kBAAkB;;iBAEtB,CAAC;AAEH,KAAK,YAAY,GAAG,CAAC,CAAC,KAAK,CAAC,OAAO,kBAAkB,CAAC,CAAC;AAEvD,wBAAgB,gBAAgB,CAAC,QAAQ,EAAE,WAAW,GAAG,OAAO,CAAC,YAAY,CAAC,CAe7E;AAED,MAAM,WAAW,iBAAiB;IAChC;;;OAGG;IACH,YAAY,CAAC,EAAE,MAAM,EAAE,CAAC;IACxB,yDAAyD;IACzD,SAAS,CAAC,EAAE,OAAO,CAAC;CACrB;AAiCD,wBAAgB,gBAAgB,CAC9B,QAAQ,EAAE,WAAW,EACrB,OAAO,GAAE,iBAAsB,GAC9B,OAAO,CAAC,YAAY,CAAC,CAgBvB;AAED,wBAAgB,QAAQ,CACtB,QAAQ,EAAE,WAAW,EACrB,aAAa,CAAC,EAAE,iBAAiB,GAChC,CAAC,OAAO,CAAC,YAAY,CAAC,EAAE,OAAO,CAAC,YAAY,CAAC,CAAC,CAEhD"}
|
package/dist/tools.js
CHANGED
|
@@ -1,11 +1,13 @@
|
|
|
1
|
+
import { realpath } from 'node:fs/promises';
|
|
2
|
+
import { isAbsolute, relative, resolve } from 'node:path';
|
|
1
3
|
import { z } from 'zod';
|
|
2
4
|
const SearchParamsSchema = z.object({
|
|
3
|
-
query: z.string().min(1),
|
|
4
|
-
limit: z.number().int().positive().optional(),
|
|
5
|
-
threshold: z.number().min(0).max(1).optional(),
|
|
5
|
+
query: z.string().min(1).describe('Search query for the knowledge base'),
|
|
6
|
+
limit: z.number().int().positive().optional().describe('Maximum number of results to return'),
|
|
7
|
+
threshold: z.number().min(0).max(1).optional().describe('Minimum relevance score (0-1)'),
|
|
6
8
|
});
|
|
7
9
|
const IngestParamsSchema = z.object({
|
|
8
|
-
source: z.string().min(1),
|
|
10
|
+
source: z.string().min(1).describe('File path, directory or URL to ingest'),
|
|
9
11
|
});
|
|
10
12
|
export function createSearchTool(pipeline) {
|
|
11
13
|
return {
|
|
@@ -24,7 +26,36 @@ export function createSearchTool(pipeline) {
|
|
|
24
26
|
},
|
|
25
27
|
};
|
|
26
28
|
}
|
|
27
|
-
|
|
29
|
+
function isUrl(source) {
|
|
30
|
+
return /^[a-z][a-z0-9+.-]*:\/\//i.test(source);
|
|
31
|
+
}
|
|
32
|
+
async function canonicalPath(path) {
|
|
33
|
+
try {
|
|
34
|
+
return await realpath(path);
|
|
35
|
+
}
|
|
36
|
+
catch {
|
|
37
|
+
return resolve(path);
|
|
38
|
+
}
|
|
39
|
+
}
|
|
40
|
+
async function assertAllowedSource(source, options) {
|
|
41
|
+
if (isUrl(source)) {
|
|
42
|
+
if (options.allowUrls === false) {
|
|
43
|
+
throw new Error('URL sources are not allowed');
|
|
44
|
+
}
|
|
45
|
+
return;
|
|
46
|
+
}
|
|
47
|
+
if (!options.allowedRoots)
|
|
48
|
+
return;
|
|
49
|
+
const target = await canonicalPath(source);
|
|
50
|
+
for (const root of options.allowedRoots) {
|
|
51
|
+
const canonicalRoot = await canonicalPath(root);
|
|
52
|
+
const rel = relative(canonicalRoot, target);
|
|
53
|
+
if (rel === '' || (!rel.startsWith('..') && !isAbsolute(rel)))
|
|
54
|
+
return;
|
|
55
|
+
}
|
|
56
|
+
throw new Error(`Source "${source}" is outside the allowed directories`);
|
|
57
|
+
}
|
|
58
|
+
export function createIngestTool(pipeline, options = {}) {
|
|
28
59
|
return {
|
|
29
60
|
name: 'rag_ingest',
|
|
30
61
|
description: 'Ingest documents into the knowledge base',
|
|
@@ -32,6 +63,7 @@ export function createIngestTool(pipeline) {
|
|
|
32
63
|
execute: async (params) => {
|
|
33
64
|
try {
|
|
34
65
|
const { source } = IngestParamsSchema.parse(params);
|
|
66
|
+
await assertAllowedSource(source, options);
|
|
35
67
|
const { documents, chunks } = await pipeline.ingest(source);
|
|
36
68
|
return { success: true, source, documents, chunks };
|
|
37
69
|
}
|
|
@@ -41,7 +73,7 @@ export function createIngestTool(pipeline) {
|
|
|
41
73
|
},
|
|
42
74
|
};
|
|
43
75
|
}
|
|
44
|
-
export function ragTools(pipeline) {
|
|
45
|
-
return [createSearchTool(pipeline), createIngestTool(pipeline)];
|
|
76
|
+
export function ragTools(pipeline, ingestOptions) {
|
|
77
|
+
return [createSearchTool(pipeline), createIngestTool(pipeline, ingestOptions)];
|
|
46
78
|
}
|
|
47
79
|
//# sourceMappingURL=tools.js.map
|
package/dist/tools.js.map
CHANGED
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"tools.js","sourceRoot":"","sources":["../src/tools.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAUxB,MAAM,kBAAkB,GAAG,CAAC,CAAC,MAAM,CAAC;IAClC,KAAK,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC;
|
|
1
|
+
{"version":3,"file":"tools.js","sourceRoot":"","sources":["../src/tools.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,QAAQ,EAAE,MAAM,kBAAkB,CAAC;AAC5C,OAAO,EAAE,UAAU,EAAE,QAAQ,EAAE,OAAO,EAAE,MAAM,WAAW,CAAC;AAC1D,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAUxB,MAAM,kBAAkB,GAAG,CAAC,CAAC,MAAM,CAAC;IAClC,KAAK,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,QAAQ,CAAC,qCAAqC,CAAC;IACxE,KAAK,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAAC,qCAAqC,CAAC;IAC7F,SAAS,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,QAAQ,EAAE,CAAC,QAAQ,CAAC,+BAA+B,CAAC;CACzF,CAAC,CAAC;AAIH,MAAM,kBAAkB,GAAG,CAAC,CAAC,MAAM,CAAC;IAClC,MAAM,EAAE,CAAC,CAAC,MAAM,EAAE,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,QAAQ,CAAC,uCAAuC,CAAC;CAC5E,CAAC,CAAC;AAIH,MAAM,UAAU,gBAAgB,CAAC,QAAqB;IACpD,OAAO;QACL,IAAI,EAAE,YAAY;QAClB,WAAW,EAAE,iDAAiD;QAC9D,UAAU,EAAE,kBAAkB;QAC9B,OAAO,EAAE,KAAK,EAAE,MAAM,EAAE,EAAE;YACxB,IAAI,CAAC;gBACH,MAAM,EAAE,KAAK,EAAE,KAAK,EAAE,SAAS,EAAE,GAAG,kBAAkB,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;gBACrE,MAAM,OAAO,GAAG,MAAM,QAAQ,CAAC,KAAK,CAAC,KAAK,EAAE,EAAE,IAAI,EAAE,KAAK,EAAE,SAAS,EAAE,CAAC,CAAC;gBACxE,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE,KAAK,EAAE,OAAO,EAAE,KAAK,EAAE,OAAO,CAAC,MAAM,EAAE,CAAC;YAClE,CAAC;YAAC,OAAO,GAAG,EAAE,CAAC;gBACb,OAAO,EAAE,OAAO,EAAE,KAAK,EAAE,KAAK,EAAE,GAAG,YAAY,KAAK,CAAC,CAAC,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,GAAG,CAAC,EAAE,CAAC;YACrF,CAAC;QACH,CAAC;KACF,CAAC;AACJ,CAAC;AAYD,SAAS,KAAK,CAAC,MAAc;IAC3B,OAAO,0BAA0B,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;AACjD,CAAC;AAED,KAAK,UAAU,aAAa,CAAC,IAAY;IACvC,IAAI,CAAC;QACH,OAAO,MAAM,QAAQ,CAAC,IAAI,CAAC,CAAC;IAC9B,CAAC;IAAC,MAAM,CAAC;QACP,OAAO,OAAO,CAAC,IAAI,CAAC,CAAC;IACvB,CAAC;AACH,CAAC;AAED,KAAK,UAAU,mBAAmB,CAAC,MAAc,EAAE,OAA0B;IAC3E,IAAI,KAAK,CAAC,MAAM,CAAC,EAAE,CAAC;QAClB,IAAI,OAAO,CAAC,SAAS,KAAK,KAAK,EAAE,CAAC;YAChC,MAAM,IAAI,KAAK,CAAC,6BAA6B,CAAC,CAAC;QACjD,CAAC;QACD,OAAO;IACT,CAAC;IAED,IAAI,CAAC,OAAO,CAAC,YAAY;QAAE,OAAO;IAElC,MAAM,MAAM,GAAG,MAAM,aAAa,CAAC,MAAM,CAAC,CAAC;IAC3C,KAAK,MAAM,IAAI,IAAI,OAAO,CAAC,YAAY,EAAE,CAAC;QACxC,MAAM,aAAa,GAAG,MAAM,aAAa,CAAC,IAAI,CAAC,CAAC;QAChD,MAAM,GAAG,GAAG,QAAQ,CAAC,aAAa,EAAE,MAAM,CAAC,CAAC;QAC5C,IAAI,GAAG,KAAK,EAAE,IAAI,CAAC,CAAC,GAAG,CAAC,UAAU,CAAC,IAAI,CAAC,IAAI,CAAC,UAAU,CAAC,GAAG,CAAC,CAAC;YAAE,OAAO;IACxE,CAAC;IACD,MAAM,IAAI,KAAK,CAAC,WAAW,MAAM,sCAAsC,CAAC,CAAC;AAC3E,CAAC;AAED,MAAM,UAAU,gBAAgB,CAC9B,QAAqB,EACrB,UAA6B,EAAE;IAE/B,OAAO;QACL,IAAI,EAAE,YAAY;QAClB,WAAW,EAAE,0CAA0C;QACvD,UAAU,EAAE,kBAAkB;QAC9B,OAAO,EAAE,KAAK,EAAE,MAAM,EAAE,EAAE;YACxB,IAAI,CAAC;gBACH,MAAM,EAAE,MAAM,EAAE,GAAG,kBAAkB,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;gBACpD,MAAM,mBAAmB,CAAC,MAAM,EAAE,OAAO,CAAC,CAAC;gBAC3C,MAAM,EAAE,SAAS,EAAE,MAAM,EAAE,GAAG,MAAM,QAAQ,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC;gBAC5D,OAAO,EAAE,OAAO,EAAE,IAAI,EAAE,MAAM,EAAE,SAAS,EAAE,MAAM,EAAE,CAAC;YACtD,CAAC;YAAC,OAAO,GAAG,EAAE,CAAC;gBACb,OAAO,EAAE,OAAO,EAAE,KAAK,EAAE,KAAK,EAAE,GAAG,YAAY,KAAK,CAAC,CAAC,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,GAAG,CAAC,EAAE,CAAC;YACrF,CAAC;QACH,CAAC;KACF,CAAC;AACJ,CAAC;AAED,MAAM,UAAU,QAAQ,CACtB,QAAqB,EACrB,aAAiC;IAEjC,OAAO,CAAC,gBAAgB,CAAC,QAAQ,CAAC,EAAE,gBAAgB,CAAC,QAAQ,EAAE,aAAa,CAAC,CAAC,CAAC;AACjF,CAAC"}
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@cogitator-ai/rag",
|
|
3
|
-
"version": "0.
|
|
3
|
+
"version": "0.3.0",
|
|
4
4
|
"description": "Retrieval Augmented Generation for Cogitator AI agents",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"main": "./dist/index.js",
|
|
@@ -14,13 +14,16 @@
|
|
|
14
14
|
"files": [
|
|
15
15
|
"dist"
|
|
16
16
|
],
|
|
17
|
+
"engines": {
|
|
18
|
+
"node": ">=22.12.0"
|
|
19
|
+
},
|
|
17
20
|
"dependencies": {
|
|
18
|
-
"
|
|
19
|
-
"
|
|
20
|
-
"
|
|
21
|
+
"@cogitator-ai/types": "0.24.0",
|
|
22
|
+
"nanoid": "^6.0.1",
|
|
23
|
+
"zod": "^4.6.5"
|
|
21
24
|
},
|
|
22
25
|
"peerDependencies": {
|
|
23
|
-
"@cogitator-ai/memory": "0.
|
|
26
|
+
"@cogitator-ai/memory": ">=0.8.0"
|
|
24
27
|
},
|
|
25
28
|
"peerDependenciesMeta": {
|
|
26
29
|
"@cogitator-ai/memory": {
|
|
@@ -29,15 +32,14 @@
|
|
|
29
32
|
},
|
|
30
33
|
"optionalDependencies": {
|
|
31
34
|
"cheerio": "^1.0.0",
|
|
32
|
-
"papaparse": "^5.
|
|
33
|
-
"pdf-parse": "^
|
|
35
|
+
"papaparse": "^5.7.0",
|
|
36
|
+
"pdf-parse": "^2.4.5"
|
|
34
37
|
},
|
|
35
38
|
"devDependencies": {
|
|
39
|
+
"@cogitator-ai/memory": "0.8.0",
|
|
36
40
|
"@types/papaparse": "^5.3.15",
|
|
37
|
-
"
|
|
38
|
-
"
|
|
39
|
-
"vitest": "^4.0.18",
|
|
40
|
-
"@cogitator-ai/memory": "0.6.22"
|
|
41
|
+
"typescript": "^6.0.3",
|
|
42
|
+
"vitest": "^5.0.3"
|
|
41
43
|
},
|
|
42
44
|
"repository": {
|
|
43
45
|
"type": "git",
|
|
@@ -45,12 +47,11 @@
|
|
|
45
47
|
"directory": "packages/rag"
|
|
46
48
|
},
|
|
47
49
|
"publishConfig": {
|
|
48
|
-
"access": "public"
|
|
49
|
-
"registry": "https://npm.pkg.github.com"
|
|
50
|
+
"access": "public"
|
|
50
51
|
},
|
|
51
52
|
"license": "MIT",
|
|
52
53
|
"scripts": {
|
|
53
|
-
"build": "tsc",
|
|
54
|
+
"build": "tsc && tsx ../../scripts/fix-esm-imports.ts .",
|
|
54
55
|
"dev": "tsc --watch",
|
|
55
56
|
"clean": "rm -rf dist",
|
|
56
57
|
"typecheck": "tsc --noEmit",
|