Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining Wiki.
Go agentagenticaichatbotdsh-pluginembeddingsevaluationgenerative-aigolangknowledge-basellmmulti-tenantollamaopenaiquestion-answeringragrerankingsemantic-searchvector-searchwiki
import assert from 'node:assert/strict'
import test from 'node:test'
import { getAgentToolIconName } from './agent-tool-icons.ts'
import {
getKnowledgeSearchSummaryHtml,
getQueryText,
getRagPipelineStepTitle,
getWikiPageText,
} from './agent-tool-display.ts'
const t = (key, params) => {
if (key === 'agentStream.search.foundResultsFromFiles') {
return `found ${params?.count} from ${params?.files} files`
}
if (key === 'agentStream.search.candidatesBelowThreshold') {
return `matched ${params?.count}, none relevant`
}
if (key === 'agentStream.ragPipeline.searchingWithQuery') {
return `searching ${params?.query}`
}
return key
}
test('getAgentToolIconName maps rag pipeline tools', () => {
assert.equal(getAgentToolIconName('query_understand'), 'ai-search')
assert.equal(getAgentToolIconName('knowledge_search'), 'data-search')
})
test('getAgentToolIconName maps the consolidated knowledge tools and their retired names alike', () => {
assert.equal(getAgentToolIconName('search_knowledge'), 'data-search')
assert.equal(getAgentToolIconName('search_knowledge', 'web'), 'internet')
assert.equal(getAgentToolIconName('read_document'), 'file-search')
assert.equal(getAgentToolIconName('list_documents'), 'file-search')
assert.equal(getAgentToolIconName('list_knowledge_chunks'), 'file-search')
assert.equal(getAgentToolIconName('get_document_info'), 'file-search')
})
test('getAgentToolIconName maps sandbox shell tools to the terminal icon', () => {
assert.equal(getAgentToolIconName('shell_exec'), 'terminal')
})
test('getAgentToolIconName maps skill and sandbox file tools', () => {
assert.equal(getAgentToolIconName('read_file'), 'file')
assert.equal(getAgentToolIconName('read_skill'), 'file')
assert.equal(getAgentToolIconName('list_sandbox_files'), 'folder')
assert.equal(getAgentToolIconName('read_sandbox_file'), 'file')
assert.equal(getAgentToolIconName('execute_skill_script'), 'code')
})
test('getAgentToolIconName maps operation and product-doc tools', () => {
assert.equal(getAgentToolIconName('find_operations'), 'task')
assert.equal(getAgentToolIconName('run_operation'), 'terminal')
assert.equal(getAgentToolIconName('search_weknora_docs'), 'search')
assert.equal(getAgentToolIconName('read_weknora_doc'), 'file')
})
test('getAgentToolIconName maps Wiki tools to semantic search and reading icons', () => {
assert.equal(getAgentToolIconName('wiki_search'), 'search')
assert.equal(getAgentToolIconName('wiki_read_page'), 'file-search')
assert.equal(getAgentToolIconName('wiki_read_source_doc'), 'file-search')
})
test('getQueryText joins unique query strings', () => {
assert.equal(getQueryText({ query: 'foo', queries: ['foo', 'bar'] }), 'foo,bar')
})
test('getQueryText parses JSON-encoded queries string', () => {
assert.equal(
getQueryText({
queries: '["合力天胜游泳俱乐部介绍", "合力天胜游泳训练机构", "合力天胜游泳队"]',
}),
'合力天胜游泳俱乐部介绍,合力天胜游泳训练机构,合力天胜游泳队',
)
})
test('getWikiPageText supports persisted slugs arrays', () => {
assert.equal(
getWikiPageText({ slugs: ['entity/知识助理', 'concept/API管理'] }),
'entity/知识助理、concept/API管理',
)
assert.equal(getWikiPageText('{"slug":"index"}'), 'index')
})
test('getKnowledgeSearchSummaryHtml includes file count when present', () => {
const html = getKnowledgeSearchSummaryHtml(t, {
results: [{}, {}],
kb_counts: { a: 1, b: 2 },
})
assert.match(html, /found <strong>2<\/strong> from <strong>2<\/strong> files/)
})
// Candidates that all fell below the relevance threshold never reached the
// answer, so the row must not read like a plain empty search: the difference
// points at the threshold rather than at the knowledge base.
test('getKnowledgeSearchSummaryHtml distinguishes filtered candidates from an empty search', () => {
assert.match(
getKnowledgeSearchSummaryHtml(t, { count: 0, candidate_count: 10 }),
/matched <strong>10<\/strong>, none relevant/,
)
assert.equal(
getKnowledgeSearchSummaryHtml(t, { count: 0, candidate_count: 0 }),
'agentStream.search.noResults',
)
})
test('getRagPipelineStepTitle uses query-aware search labels', () => {
const title = getRagPipelineStepTitle(t, {
tool_name: 'knowledge_search',
pending: true,
arguments: { query: '讯飞开放平台' },
})
assert.equal(title, 'searching 讯飞开放平台')
})
test('getRagPipelineStepTitle uses web labels when search_source is web', () => {
const title = getRagPipelineStepTitle(t, {
tool_name: 'knowledge_search',
pending: false,
success: true,
arguments: { search_source: 'web' },
})
assert.equal(title, 'agentStream.toolStatus.webSearch')
})
test('getRagPipelineStepTitle uses attachment parsing labels', () => {
assert.equal(
getRagPipelineStepTitle(t, { tool_name: 'attachment_parsing', pending: true }),
'agentStream.toolStatus.attachmentParsing',
)
assert.equal(
getRagPipelineStepTitle(t, { tool_name: 'attachment_parsing', pending: false, success: true }),
'agentStream.toolStatus.attachmentParsingDone',
)
})