Research conclusions
- Discovery state should be deterministic and server-owned; the model owns phrasing, not truth.
- Score access persistence and content sensitivity separately.
- Use explicit qualify, prototype, disqualify and escalate terminal states.
P01 · Demand knowledge node
Client intelligence and discovery
Begin with everything Actionist already knows, enrich it with bounded public business context, then ask only the questions needed to identify valuable workflows.
What we learned
Current synthesis from the research packet. Exact records and complete source artifacts remain visible below.
What it owns
- Known client/account context
- Business links and bounded site reconnaissance
- Stakeholders, systems of record and constraints
- Conversation history and unresolved discovery questions
What research already tells us
- Actionist can pre-populate client context before chat
- A generic “what app do you want?” prompt wastes known information
- The useful target is a painful workflow and measurable outcome, not a feature wishlist
Open first-principles questions
- Which Actionist client fields are canonical?
- What may be scraped automatically?
- How do we distinguish stated wants from observed workflow pain?
- What confidence threshold triggers a follow-up question?
Inputs
- Actionist account profile
- Client URLs
- Conversation
- Existing tools
Outputs
- ClientContext
- DiscoveryGapList
- CandidateOutcomes
Agent-sized workstream
Map Actionist client context, define the discovery schema, question-selection policy and confidence model.
Linked research
- Industry atom specifications
research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md - Actionist solutions sweep specification
research/actionist-solutions-sweep-spec-2026-08-26.md - Public signals Wave 11
research/actionmodel-builder-research-2026-08-26/expansion/wave-11/outputs/public-signals-wave-11.md
Connected knowledge nodes
Where this beats a generator
Lovable starts every project from an empty prompt box, so the client must already know what to ask for. Starting from known account context and public business signals means the first question is about a painful workflow, not a feature wishlist.
Companies and commercial surfaces
55 mined records. Search and inspect the full agent output; nothing is collapsed into a top-ten headline.
| Rank | Identity | What the agent found | Evidence | Source |
|---|---|---|---|---|
| — | Auctorai-discovery | AI agents join enterprise software implementation discovery and refinement calls, capture requirements as they surface, and generate SOWs, proposals, architecture docs, user stories, test plans and configurations from them. | Drelevant | Source ↗ |
| — | Cloobot Xai-discovery | A 'CRM for Requirements' that converts workshops, stakeholder conversations, notes and SoWs into structured requirements with live traceability from requirement to user story to Jira ticket to release. | Drelevant | Source ↗ |
| — | Fin for Sales (Intercom)ai-discovery | An inbound AI SDR that runs sales conversations end to end — engaging visitors, guiding discovery on pricing/features/objections, qualifying against a playbook, and booking a meeting, starting a trial, or handing off with full context. | Drelevant | Source ↗ |
| — | Avomaconversation-intelligence | All-in-one AI meeting assistant and conversation/revenue intelligence platform that records, transcribes, and structures meetings, and automatically maps meeting content into sales-methodology CRM fields. | Drelevant | Source ↗ |
| — | Gongconversation-intelligence | Revenue AI platform that automatically captures every customer interaction, maps it to business context in a 'Revenue Graph', and surfaces deal risk and opportunity insights. | Drelevant | Source ↗ |
| — | Lawmaticsintake | AI-powered legal CRM whose intake product turns inquiries into booked consults and signed agreements, with automatic conflict scanning and AI lead scoring. | Drelevant | Source ↗ |
| — | Loopiointake | RFP, DDQ and security-questionnaire response platform that auto-populates answers from a curated content Library and scores every AI-generated answer for confidence. | Drelevant | Source ↗ |
| — | Conveyorintake | Security-questionnaire automation that auto-completes complex questionnaires by pulling from existing documents, wikis and external sites rather than requiring a manually curated answer library. | Irelevant | Source ↗ |
| — | Zoovuguided-selling | AI product-discovery platform unifying search, guided-selling assistants, product finders, quizzes and CPQ/configuration on one engine for B2C and B2B ecommerce. | Drelevant | Source ↗ |
| — | excentosguided-selling | Guided-selling software (Product Finder, Product Configurator, Set Builder, Troubleshooting Guide) blending traditional guided selling with generative AI to act as a digital advisor. | Drelevant | Source ↗ |
| — | Ignitionintake | Proposal-to-payment platform for professional services: online proposals that double as engagement agreements, capturing payment details at signature and automating billing thereafter. | Drelevant | Source ↗ |
| — | Karbonintake | Accounting practice-management platform combining workflow automation, email triage, client collaboration and an AI coworker (Kai), with onboarding delivered as templated workflows and client requests. | Dadjacent | Source ↗ |
| — | Demandbaseenrichment | Account intelligence platform that identifies anonymous website visitors as companies, enriches CRM records at scale, and combines firmographic, technographic, buyer and intent data into an account view. | Drelevant | Source ↗ |
| — | HubSpot Breeze Intelligence (formerly Clearbit)enrichment | Native HubSpot enrichment layer, the absorbed remains of Clearbit, offering firmographic enrichment and smart form shortening inside the CRM. | Irelevant | Source ↗ |
| — | Clayenrichment | GTM data platform running waterfall enrichment across 100+ data providers, with an AI research agent (Claygent) that browses the web to answer custom research questions per account. | Irelevant | Source ↗ |
| — | Apollo.ioenrichment | Self-serve B2B database and sales-engagement platform with an enrichment API, claiming 230M+ contacts and 30M+ companies. | Iadjacent | Source ↗ |
| — | ZoomInfoenrichment | Largest B2B database (claimed 500M+ contacts) with a company enrichment API returning firmographics, revenue, headcount, location, industry codes and social presence from a domain, name or ZoomInfo ID in one call. | Iadjacent | Source ↗ |
| — | Chorus by ZoomInfoconversation-intelligence | Conversation intelligence enriched with ZoomInfo buyer intent and firmographic data, adding context about who is on the call directly into call analysis. | Iadjacent | Source ↗ |
| — | Clari (incl. Clari Copilot / Wingman)conversation-intelligence | Revenue platform centred on forecasting and pipeline inspection, with Clari Copilot (formerly Wingman) providing conversation intelligence and real-time call guidance. | Iadjacent | Source ↗ |
| — | Salesloftconversation-intelligence | Revenue orchestration platform with sales engagement, deal management, forecasting and conversation intelligence. | Iweak | Source ↗ |
| — | Fireflies.aiconversation-intelligence | Meeting recorder and AI notetaker with the broadest platform compatibility (Zoom, Teams, Google Meet, RingCentral, Webex, uploaded audio) and transparent tiered pricing including a free plan. | Iadjacent | Source ↗ |
| — | Jiminnyconversation-intelligence | Conversation intelligence and coaching platform positioned for mid-market revenue teams. | Iadjacent | Source ↗ |
| — | Revenue.ioconversation-intelligence | Salesforce-centric conversation intelligence with real-time coaching during live calls and automated methodology scoring. | Iadjacent | Source ↗ |
| — | Clio Growintake | Legal practice management suite with client intake forms, scheduling and onboarding integrated into matter management. | Irelevant | Source ↗ |
| — | PracticePantherintake | Legal practice management with customisable intake forms, e-signatures and automated client onboarding. | Iadjacent | Source ↗ |
| — | Leglintake | UK legal client onboarding and lifecycle platform combining automated onboarding with built-in KYC, KYB and AML compliance checks. | Irelevant | Source ↗ |
| — | Content Snareintake | Client information and document collection tool built specifically for the problem of chasing clients for content, assets and answers. | Irelevant | Source ↗ |
| — | Dubsadointake | Client management platform for freelancers and small agencies automating the full intake sequence: forms, proposals, contracts, invoicing and workflows. | Irelevant | Source ↗ |
| — | HoneyBookintake | All-in-one clientflow platform for independent service businesses covering inquiry capture, proposals, contracts, payments and scheduling. | Iweak | Source ↗ |
| — | Bonsaiintake | Freelancer/agency business platform bundling proposals, attorney-vetted contract templates, invoicing, accounting and tax tools. | Iweak | Source ↗ |
| — | Financial Centsintake | Accounting practice management platform with client onboarding via workflow templates and automated client document chasing. | Dadjacent | Source ↗ |
| — | TaxDomeintake | Accounting practice management with client portal, organisers, e-signature and pipeline-based onboarding. | Irelevant | Source ↗ |
| — | Canopyintake | Accounting practice management suite with client portal, document management and engagement workflows. | Iweak | Source ↗ |
| — | Moxointake | Client interaction and workflow automation platform used for onboarding in legal, financial services and professional services, coordinating tasks and document exchange in a shared workspace. | Iadjacent | Source ↗ |
| — | Motion.iointake | Client onboarding and collaboration platform for agencies and firms, automating the client journey through portals and workflows. | Iadjacent | Source ↗ |
| — | Assemblyintake | White-label client portal software serving law firms, accounting and bookkeeping practices, marketing agencies and consultancies, with secure file sharing and automations. | Iweak | Source ↗ |
| — | Osprey Approachintake | UK legal practice suite combining case management, practice management, legal accounting and client onboarding with digital identity verification. | Iweak | Source ↗ |
| — | Clientmanager.iointake | Client onboarding and management tool aimed at agencies, covering onboarding checklists and client tracking. | Iweak | Source ↗ |
| — | Plutiointake | All-in-one business management tool for freelancers covering projects, proposals, invoices and client onboarding. | Iweak | Source ↗ |
| — | Responsive (formerly RFPIO)intake | Response management platform for high-volume RFPs and security questionnaires, with AI-assisted drafting from a centralised content library and workflow governance across concurrent projects. | Iadjacent | Source ↗ |
| — | Vantaintake | Compliance automation platform that bundles a trust centre and security questionnaire support alongside continuous control monitoring. | Iadjacent | Source ↗ |
| — | Drataintake | Compliance automation with continuous control monitoring and trust centre, competing directly with Vanta. | Iweak | Source ↗ |
| — | SafeBaseintake | Trust-centre-first platform that pre-empts security questionnaires by publishing a self-serve security posture page. | Iadjacent | Source ↗ |
| — | Whisticintake | Vendor security assessment and trust-profile platform for both sending and answering security assessments. | Iweak | Source ↗ |
| — | Outreach (AI Prospecting Agent)ai-discovery | Sales execution platform whose AI Prospecting Agent researches accounts, generates messaging, prioritises targets and enrols prospects into sequences. | Iadjacent | Source ↗ |
| — | Persana AIai-discovery | Autonomous AI sales agent managing outreach from lead discovery to meeting booking, finding qualified leads and personalising with live data. | Iweak | Source ↗ |
| — | SellScale / Selixai-discovery | AI sales agent (Selix) that posts updates in Slack explaining what it is researching and what it discovered, and proposes actions for human review before execution. | Drelevant | Source ↗ |
| — | MindStudioai-discovery | AI agent building platform; publishes an explicit discovery-question methodology for scoping and pricing AI automation engagements, and ships a product-manager agent ('Remy'). | Dadjacent | Source ↗ |
| — | Inleadoenrichment | Purpose-built inbound demo-request enrichment: prospect supplies name and business email, the platform extracts the domain, appends revenue, employees, industry and location in milliseconds, then routes by fit. | Irelevant | Source ↗ |
| — | Exploriumenrichment | External data platform with a documented Firmographics enrichment API endpoint returning revenue, company size, industry classification and other attributes. | Dadjacent | Source ↗ |
| — | Cognismenrichment | European-focused B2B data platform with GDPR-structural compliance, DNC screening across 12 EU countries, and phone-verified contact data. | Iweak | Source ↗ |
| — | CanvasLogicguided-selling | CPQ combining configuration logic with 3D visual product configuration for modular, variant-rich manufactured products. | Iadjacent | Source ↗ |
| — | PROS Smart CPQguided-selling | CPQ with an AI pricing and recommendation engine that analyses past sales records and market data to identify optimal pricing, upsell and bundling. | Iadjacent | Source ↗ |
| — | SAP CPQguided-selling | Enterprise CPQ supporting user-guided sales processes, multi-level approvals and advanced pricing structures for complex assembly configuration. | Iweak | Source ↗ |
| — | Outfindoguided-selling | Buyer-facing product-selection guide for ecommerce and self-serve B2B, converting product complexity into a short question flow. | Iadjacent | Source ↗ |
Repositories and OSS precedents
50 mined records. Search and inspect the full agent output; nothing is collapsed into a top-ten headline.
| Rank | Identity | What the agent found | Evidence | Source |
|---|---|---|---|---|
| — | process-intelligence-solutions/pm4pyprocess-mining | Reference Python process-mining library: discovery (Alpha/Heuristics/Inductive miner), conformance checking, XES + OCEL 2.0 event logs, Petri nets/BPMN/process trees. | Drelevant | Source ↗ |
| — | apromore/ApromoreCoreprocess-mining | Open-source Core Edition of the Apromore process-mining platform: log/BPMN importer, portal, Process Discoverer, business calendar, model editor. | Dweak | Source ↗ |
| — | bupaverse/bupaRprocess-mining | Core R package of the bupaR ecosystem for business process analysis; event-log object model, case/trace statistics, companion packages for discovery and performance analysis. | Drelevant | Source ↗ |
| — | bupaverse/pm4pyprocess-mining | R interface bridging bupaR to the Python PM4Py library; exposes inductive miner discovery, Petri net rendering and conformance alignments to R users. | Iadjacent | Source ↗ |
| — | computertechworld/ProcessMiningprocess-mining | Curated list of open-source process-mining tools and resources (ProM, Apromore CE, RapidProM, OpenXES). | Iadjacent | Source ↗ |
| — | TheWoops/awesome-processminingprocess-mining | Collection of process-mining resources. | Iweak | Source ↗ |
| — | RPA-US/screenrpatask-mining | Screen-based task-mining framework for RPA: Django platform capturing UI logs, eye-tracking gaze logs and screenshots, with modules for behaviour monitoring, decision discovery, feature extraction, process discovery and reporting. | Dweak | Source ↗ |
| — | apromore/RPA_UILoggertask-mining | User-interaction logger for robotic process mining; Excel add-in passive listener plus Chrome extension, writing captured UI events to CSV. | Dadjacent | Source ↗ |
| — | OpenAdaptAI/OpenAdapttask-mining | Compiles a demonstrated GUI workflow into a deterministic, locally executable program; captures structural, accessibility, visual, OCR, spatial and transition evidence across browser, Windows, macOS, Linux, RDP and Citrix/VDI. | Drelevant | Source ↗ |
| — | xlang-ai/OpenCUAtask-mining | NeurIPS 2025 spotlight framework for computer-use agents: AgentNetTool cross-platform recorder, AgentNet dataset of 22.6K human-annotated desktop tasks across 200+ apps, AgentNetBench evaluator, OpenCUA 7B/32B/72B models. | Drelevant | Source ↗ |
| — | ActivityWatch/activitywatchtask-mining | Cross-platform automated time tracker; aw-watcher-window logs focused application and window title, aw-watcher-afk infers active/idle from keyboard and mouse, aw-watcher-web browser extension tracks active tab title and URL. | Drelevant | Source ↗ |
| — | humblebanana/open-record-replaytask-mining | macOS record-and-replay workflow recorder for computer-use agents; captures mouse, keyboard and UI events as structured traces (session.json, events.jsonl), validates recording quality and packages evidence for an agent skill-creation flow. | Iadjacent | Source ↗ |
| — | A9T9/RPAtask-mining | Ui.Vision open-source RPA with computer vision, OCR, Anthropic Computer Use/LLM integration and a web macro recorder; Selenium IDE import/export. | Iweak | Source ↗ |
| — | aisingapore/TagUItask-mining | Free RPA tool by AI Singapore covering websites, desktop apps and CLI; cross-platform; flows writable in 22 languages. | Iweak | Source ↗ |
| — | robocorp/rpaframeworktask-mining | Collection of open-source RPA libraries for Robot Framework and Python. | Iweak | Source ↗ |
| — | firecrawl/fire-enrichenrichment | Multi-agent enrichment pipeline turning an email/domain into a company profile: industry, business model, employee count, founding year, HQ, funding stage and investors, tech stack, plus custom fields — every field carrying source URLs. | Drelevant | Source ↗ |
| — | Lissy93/web-checkenrichment | All-in-one website OSINT tool running ~40 checks: DNS, SSL/TLS config and cipher suites, HTTP and security headers, cookies, Whois, robots.txt, sitemap/listed pages, open ports, traceroute, trackers, tech stack via Wappalyzer fingerprints, email config (DMARC/DKIM/SPF/BIMI), WAF detection, Wayback archive history, Tranco global rank, screenshots. | Drelevant | Source ↗ |
| — | specfy/stack-analyserenrichment | Extracts 700+ technologies from a repository — languages, SaaS products, cloud, infrastructure, dependencies and services — by parsing package.json, docker-compose.yml, go.mod and similar manifests, and mapping folder-to-service relationships. | Drelevant | Source ↗ |
| — | mattvinall/Quick-Enrich-Toolsenrichment | Open-source suite of six lead-enrichment tools intended to be cloned, customized and self-hosted with your own API keys; reference stack Railway + Vercel + Supabase + Resend at roughly $10-20/mo. | Iweak | Source ↗ |
| — | eracle/OpenOutreachenrichment | Self-hosted CLI agent for B2B lead generation: describe a product, it finds matching people from a licensed data provider and explains the fit, exporting CSV. Claims zero platform-ToS surface — browserless, no social account, no scraping. | Iadjacent | Source ↗ |
| — | taranis-ai/taranis-aienrichment | AI-assisted OSINT tool for information gathering and situational analysis. | Iweak | Source ↗ |
| — | jivoi/awesome-osintenrichment | Curated list of OSINT tools and resources. | Iweak | Source ↗ |
| — | openclarity/apiclarityintrospection | Captures and analyzes API traffic to reconstruct OpenAPI specifications with no code instrumentation, then diffs observed traffic against approved specs to flag Shadow APIs (observed but undocumented) and Zombie APIs (deprecated but still live). | Dadjacent | Source ↗ |
| — | grokify/traffic2openapiintrospection | Normalizes HTTP traffic from HAR files, Postman collections, Playwright/Puppeteer/Cypress captures, mitmproxy/Charles logs and Go http.Client into a versioned intermediate representation ('ir.v1'), then infers OpenAPI 3.0/3.1/3.2 specs — detecting path/query params, types and formats, auth schemes and pagination patterns, and diffing specs for breaking changes. | Drelevant | Source ↗ |
| — | doitintl/DoiT-AdminPulse-for-Workspaceintrospection | Google Sheets add-on for Workspace admins: security-posture checklist plus read-only Admin SDK inventory scripts covering users, license assignments, Groups settings and membership, mobile devices, shared drives, OAuth tokens (getOAuthTokens.gs), app passwords, OUs, domains and DNS. | Drelevant | Source ↗ |
| — | psignoret/azure-ad-permissions-gistintrospection | Widely-used community PowerShell script enumerating all delegated and application permission grants in Entra ID, iterating app permissions directly rather than parsing logs; updated to handle tenants with 1000+ grants. | Iadjacent | Source ↗ |
| — | walidamro-msft/entra-id-graph-consent-toolkitintrospection | Entra ID governance toolkit to list, audit and bulk-delete Microsoft Graph delegated user consents, forcing least-privilege re-consent; PowerShell 5.1 and 7+, revocation path uses pure REST with no module dependency. | Iadjacent | Source ↗ |
| — | MianAzeem/api-snifferintrospection | Playwright-based tool capturing real-time API traffic from web applications and generating structured API intelligence — schema inference, automatic endpoint classification (AUTH, USER), and an OpenAPI generator. | Iadjacent | Source ↗ |
| — | NangoHQ/nangointrospection | Self-hostable product-integration platform providing managed OAuth and unified syncs across a large catalog of third-party SaaS APIs. | Iadjacent | Source ↗ |
| — | akto-api-security/aktointrospection | Open-source API security platform whose code-analysis tool scans source repositories to discover APIs including shadow APIs, integrating with traffic-based inventory and CI/CD to catch new APIs in pull requests. | Iadjacent | Source ↗ |
| — | opticdev/opticintrospection | Captures real HTTP traffic from production or elsewhere and creates OpenAPI from it, for documentation, mocks, SDKs or contract testing. | Iadjacent | Source ↗ |
| — | schemathesis/schemathesisintrospection | Property-based API testing driven by OpenAPI and GraphQL schemas. | Iweak | Source ↗ |
| — | osquery/osqueryintrospection | Exposes an operating system as a relational database, allowing SQL queries over installed applications, running processes, browser extensions and network connections. | Iadjacent | Source ↗ |
| — | zeek/zeekintrospection | Network security monitor generating structured logs of application traffic; can reveal SaaS usage by inspecting DNS queries, IP connections and TLS/SNI metadata without decrypting content. | Iadjacent | Source ↗ |
| — | modAL-python/modALactive-learning | Modular active-learning framework on scikit-learn: ActiveLearner with pluggable query strategies (uncertainty, entropy, custom), Committee/CommitteeRegressor for query-by-committee, BayesianOptimizer, ranked batch-mode sampling and information density; Keras integration. | Drelevant | Source ↗ |
| — | scikit-activeml/scikit-activemlactive-learning | Comprehensive active-learning library on scikit-learn implementing 60+ query strategies drawn from 40+ research papers, spanning pool-based and stream-based settings, with deep active learning via skorch/PyTorch. | Drelevant | Source ↗ |
| — | usnistgov/optbayesexptactive-learning | Optimal Bayesian Experimental Design for choosing measurement settings when a parametric model is known; selects settings likely to make parameter estimates more precise, aimed at cases where measurements are expensive. | Irelevant | Source ↗ |
| — | experimental-design/bofireactive-learning | Framework for experimental design and multi-objective Bayesian optimization, built on Meta's BoTorch. | Iadjacent | Source ↗ |
| — | emdgroup/baybeactive-learning | General-purpose toolbox for Bayesian Design of Experiments focused on real-world experimental campaigns; claimed use in industry and academia. | Iadjacent | Source ↗ |
| — | VlachosGroup/nextorchactive-learning | Experimental design and Bayesian optimization library in Python/PyTorch. | Iweak | Source ↗ |
| — | mhinne/BNQDactive-learning | Toolbox for Bayesian Nonparametric Quasi-Experimental Design in Python. | Iweak | Source ↗ |
| — | bayesflow-org/bayesflowactive-learning | Amortized Bayesian inference library; ships a Bayesian Experimental Design example notebook framing BED as combining information theory and Bayesian modelling to plan experiments maximizing information gain. | Iadjacent | Source ↗ |
| — | microsoft/graphragorg-graph | Modular graph-based RAG system; an LLM-driven pipeline that extracts structured entities and relationships from unstructured narrative text and builds knowledge-graph memory structures used as targeted context for question answering. | Drelevant | Source ↗ |
| — | getzep/graphitiorg-graph | Framework for real-time temporal knowledge graphs for AI agents: incremental ingestion without full recomputation, bi-temporal fact validity windows with automatic invalidation on contradiction, provenance via 'episodes' linking every fact to its source, hybrid semantic + BM25 + graph-traversal retrieval, custom ontology through Pydantic entity and edge types. | Drelevant | Source ↗ |
| — | zjunlp/DeepKEorg-graph | Open toolkit for knowledge-graph extraction and construction (EMNLP 2022): named entity recognition, relation extraction, attribute extraction, event extraction and relational triple extraction, across standard, few-shot, document-level and multimodal settings; DeepKE-LLM branch, OneKE multi-agent extraction framework, MCP service tools. | Drelevant | Source ↗ |
| — | robert-mcdermott/ai-knowledge-graphorg-graph | Takes unstructured text, uses any OpenAI-compatible LLM (Ollama, LM Studio, vLLM, LiteLLM, OpenAI) to extract subject-predicate-object triplets, standardizes entities across document chunks, infers relationships to connect disconnected subgraphs, and renders an interactive graph. | Iadjacent | Source ↗ |
| — | junhewk/simple-graph-builderorg-graph | Obsidian plugin building a lightweight knowledge graph from notes via LLM entity extraction over a fixed 10-type ontology (PERSON, ORGANIZATION, CONCEPT, PROJECT, TOOL, EVENT, PLACE, DOCUMENT, METHOD, TOPIC) with free-form relationship verbs; hybrid entity resolution resolving most entities by hash lookup and reserving embedding search and LLM calls for genuinely ambiguous cases. | Iadjacent | Source ↗ |
| — | MadsDoodle/Knowledge-Graph-using-RDF-LLMorg-graph | Pipeline extracting entities and relationships from PDF, TXT and DOCX into RDF graphs, combining SpaCy rule-based NER with LLM extraction, plus deduplication and merging; SPARQL-queryable. | Iweak | Source ↗ |
| — | dylanhogg/llmgraphorg-graph | Creates knowledge graphs in GraphML, GEXF and HTML from a source entity's Wikipedia page, with customised prompts per entity type and cache support for iteratively growing a graph. | Iweak | Source ↗ |
| — | FabioYanezRomero/Knowledge-Graph-Builderorg-graph | Builds knowledge graphs via generative models through Ollama/Gemini/LM Studio, with structured triple extraction source-grounded via langextract, graph augmentation to bridge disconnected components, schema-constrained customizable domains, Cytoscape.js visualization and GraphML output. | Iadjacent | Source ↗ |
Innovation register
50 mined records. Search and inspect the full agent output; nothing is collapsed into a top-ten headline.
| Rank | Identity | What the agent found | Evidence | Source |
|---|---|---|---|---|
| — | inn-p01-001— | Precompute-first discovery ladder: rank every ClientContext field by acquisition cost (public scrape < enrichment API < integration read < question) and never ask what a cheaper rung can supply | hypothesiscandidate | lane reasoning over parts.json P01 thesis |
| — | inn-p01-002— | Industry-prior Bayesian warm start: initialize the client model from the 17-industry prior (entities, systems, atoms, pain distribution) so the first question already discriminates within-industry variance rather than re-deriving the industry | hypothesiscandidate | industry-discovery-priors.jsonl |
| — | inn-p01-003— | Expected-value-of-information (EVOI) question selector: score each candidate question by how much its answer is expected to change the recommended workflow set, not by template order | hypothesiscandidate | lane reasoning; active-learning literature (survey pending) |
| — | inn-p01-004— | Stop asking when decision-invariant: stopping rule fires when the top-ranked CandidateOutcome is stable under all plausible answers to remaining questions | hypothesiscandidate | lane reasoning |
| — | inn-p01-005— | Wants-vs-pain dual ledger: record stated wants and observed/derived pain as separate typed streams; a recommendation must cite at least one pain evidence item, not only wants | hypothesiscandidate | parts.json P01 open question; phase-2 contradiction fields |
| — | inn-p01-006— | Contradiction as first-class object: when scraped evidence conflicts with a stated answer (e.g., stale listings vs 'all current'), persist a Contradiction record with both sources and surface it as a clarifying question rather than silently preferring either | hypothesiscandidate | phase-2 industry contradiction risks |
| — | inn-p01-007— | Stack fingerprinting from public surfaces: detect storefront/PMS/ATS/CMS/analytics from the client's public web presence (headers, scripts, job posts, DNS) to pre-fill systems-of-record before any question | hypothesiscandidate | ecommerce/saas/recruiting priors note public detectability |
| — | inn-p01-008— | Job-posting mining as workflow evidence: a client's own job ads describe their processes, tools and volumes; parse them into WorkflowEvidence with provenance | hypothesiscandidate | lane reasoning |
| — | inn-p01-009— | Question budget as explicit client-effort contract: cap discovery at N minutes/questions per session, chosen from client seniority and deal size; the selector optimizes within budget rather than asking until confident | hypothesiscandidate | lane reasoning; parts.json 'client effort' analysis axis |
| — | inn-p01-010— | Answer-reuse across engagements: tenant-scoped memory so a returning client is never asked anything already answered; deltas only (what changed since last time) | hypothesiscandidate | parts.json P01 owns conversation history |
| — | inn-p01-011— | Multi-stakeholder discovery routing: model the client as several actors (owner, ops lead, admin) and route each question to the cheapest credible answerer instead of one contact answering everything | hypothesiscandidate | phase-2 terminal-owner fields |
| — | inn-p01-012— | Show-your-model discovery UI: display the evolving ClientContext (systems, workflows, pains, confidence) to the client during discovery so they correct the model instead of answering a quiz | hypothesiscandidate | lane reasoning |
| — | inn-p01-013— | Confidence-typed ClientContext fields: every field carries {value, source, evidence_class, confidence, observed_at}; downstream P02 consumes confidence, not bare values | hypothesiscandidate | AGENTS.md evidence discipline applied to product |
| — | inn-p01-014— | Discovery from the tool the client already opens: importable evidence bundles (CSV exports, mailbox stats, calendar density) requested as optional uploads, each mapped to which questions they eliminate | hypothesiscandidate | lane reasoning; workflow-mining survey pending |
| — | inn-p01-015— | Screen-recording workflow walkthrough: client records a 5-minute 'show me how you do X today'; system extracts steps/systems/handoffs as WorkflowEvidence — task-mining without installing anything | hypothesiscandidate | task-mining segment of OSS survey (pending) |
| — | inn-p01-016— | Volume denominators before value claims: discovery must capture at least one countable denominator (tickets/week, invoices/month) per candidate workflow; a workflow without a denominator cannot be ranked | hypothesiscandidate | project evidence standard (denominator discipline) applied to discovery |
| — | inn-p01-017— | Pain-cost triangulation: estimate workflow cost from (volume denominator × time-per-item × who does it) captured as three separate low-precision answers, rather than asking 'how much does this cost you' | hypothesiscandidate | lane reasoning |
| — | inn-p01-018— | Negative-space probe: explicitly ask one question designed to falsify the current best hypothesis (disconfirmation quota) so discovery doesn't only confirm the industry prior | hypothesiscandidate | falsifiable-investigation discipline |
| — | inn-p01-019— | Authority-boundary discovery as its own track: capture who may approve sends/writes/money moves early, because authority shape changes which workflows are even proposable | hypothesiscandidate | phase-2 authority_boundary fields; atom contract authority family |
| — | inn-p01-020— | System-of-record conflict detector: when two named systems both claim an entity (CRM and spreadsheet both 'own' leads), emit a SourceOfTruthConflict requiring resolution before spec, since it predicts adaptation cost | hypothesiscandidate | phase-2 source_of_truth fields; SISOCRM one-owner rule |
| — | inn-p01-021— | Discovery-session compiler: after each conversation, compile transcript → typed ClientContext delta + unresolved DiscoveryGapList, so the artifact (not the chat log) is the state | hypothesiscandidate | conversation-intelligence segment (survey pending) |
| — | inn-p01-022— | Two-sided confidence thresholds: a follow-up question triggers when confidence is low AND the field is decision-relevant for the current candidate set; low-confidence irrelevant fields are left unknown deliberately | hypothesiscandidate | parts.json P01 open question on threshold |
| — | inn-p01-023— | Industry-prior drift telemetry: log where clients contradict the industry prior; periodically re-fit priors from accumulated engagements (P15 feedback loop applied to discovery) | hypothesiscandidate | P15 thesis joined to P01 |
| — | inn-p01-024— | Pre-meeting client brief auto-draft: before the first call, generate a one-page 'what we think we know' brief (stack, size, likely pains from prior) — turning discovery into confirmation and correction | hypothesiscandidate | pre_meeting_account_briefs use case turned inward |
| — | inn-p01-025— | Effort-aware modality mixing: same information gathered via form field, chat question, call snippet or upload; selector picks modality by client persona (desk worker vs field operator) not one-size chat | hypothesiscandidate | lane reasoning |
| — | inn-p01-026— | Regulated-industry discovery profiles: healthcare/law/mortgage get a distinct discovery mode that never ingests regulated content, asks about categories not records, and records the exclusion as evidence | hypothesiscandidate | healthcare/law priors; discovery-must-avoid-clinical-data note |
| — | inn-p01-027— | CandidateOutcome portfolio, not single answer: discovery outputs 2–4 ranked candidate outcomes with the evidence for each, and the client picks; picking is itself high-value evidence | hypothesiscandidate | parts.json P01 outputs CandidateOutcomes (plural) |
| — | inn-p01-028— | Time-boxed async discovery: send the client 3 questions/day over chat channels they already use rather than a 45-minute session; measure completion lift | hypothesiscandidate | lane reasoning on client effort |
| — | inn-p01-029— | Question phrasing bound to industry vocabulary: render each canonical question in the industry's own terms from the priors file (e.g., 'matters' not 'projects' for law) — measured effect on answer quality | hypothesiscandidate | industry priors vocabulary layer |
| — | inn-p01-030— | Explicit unknowns ship with the handoff: DiscoveryGapList is a typed artifact consumed by P02, which must convert each gap into either an assumption (labelled) or an acceptance-test question — no silent gap loss | hypothesiscandidate | parts.json P01→P02 output contract |
| — | inn-p01-031— | Client-effort ROI display: show the client 'answering these 4 questions removes 2 weeks of build risk' — making question cost/value explicit to raise completion of high-value questions | hypothesiscandidate | lane reasoning |
| — | inn-p01-032— | Integration-read consent ladder: request read-only OAuth scopes ordered by evidence value per intrusiveness (calendar metadata → mailbox counts → CRM objects), each unlocking precomputation that removes question blocks | hypothesiscandidate | introspection segment of OSS survey (pending); P11 connector plane |
| — | inn-p01-033— | Denominator-first industry scoring for pilot choice: S3-L1 needs all-17 scoring; P01 should define the score now as f(evidence availability, authority risk, workflow value density) computable from priors + discovery data | hypothesiscandidate | task-graph S3-L1 dependency |
| — | inn-p01-034— | Anti-questionnaire guarantee as product positioning: publicly commit 'we never ask more than N questions before showing you something' — a falsifiable UX contract that differentiates from agency discovery decks | hypothesiscandidate | lane reasoning; P02 prototype-vs-ask question |
| — | inn-p01-035— | Second-source rule for load-bearing fields: any field that changes the archetype choice (industry, primary system of record, volume order-of-magnitude) requires two independent sources or an explicit single-source flag | hypothesiscandidate | evidence discipline applied to ClientContext |
| — | inn-p01-036— | Discovery replay determinism: same evidence in → same ClientContext out; model-extracted fields carry extraction-version so re-running with a better model is a diff, not a rewrite | hypothesiscandidate | deterministic-spine preference in composition plane applied to demand plane |
| — | inn-p01-037— | Competitor-app teardown as demand evidence: if the client names a tool they envy, teardown its public surface into feature atoms and add to CandidateOutcomes with 'envy' provenance class | hypothesiscandidate | lane reasoning |
| — | inn-p01-038— | Cold-start question pack per industry, versioned: ship 17 curated 8-question opening packs derived from priors; every pack question must name which ClientContext fields it fills and which candidate outcomes it discriminates | hypothesiscandidate | industry priors must_ask fields |
| — | inn-p01-039— | Trust-graded evidence ingestion: client-uploaded documents are data not instructions (prompt-injection boundary); discovery pipeline strips/flags any imperative content before model exposure | hypothesiscandidate | C04 stop condition in first-principles framework |
| — | inn-p01-040— | Discovery-quality score as a gate: an engagement may not proceed to ProductSpec until ClientContext reaches a minimum completeness/confidence profile per archetype — the P01/P02 seam becomes a measurable contract | hypothesiscandidate | lane reasoning on P01→P02 contract |
| — | inn-p01-041— | Ask-the-system-not-the-person: where an integration read exists, compute the answer and ask the client only to confirm anomalies ('you seem to get ~40 tickets/week — right?') — confirmation is cheaper than recall | hypothesiscandidate | precompute ladder + conversation design |
| — | inn-p01-042— | Explicit business-loop capture for idea-type clients: when the client is starting a business (72-ideas surface), discovery must capture business loop + monetization event, not just feature wants — reusing the sweep-spec's idea decomposition rule | hypothesiscandidate | actionist-solutions-sweep-spec idea rule |
| — | inn-p01-043— | Session-zero from Actionist account data: if the client already uses Actionist agents, their existing workflow/agent/tool usage is the richest pain evidence available; mine it (with consent) before anything external | hypothesiscandidate | parts.json P01 inputs 'Actionist account profile'; P11 dependency |
| — | inn-p01-044— | Calibration ledger for the question model: every precomputed guess later confirmed/corrected by the client is logged as a calibration pair; the discovery system reports its own hit rate as a trust metric | hypothesiscandidate | lane reasoning; prove-before-claim culture |
| — | inn-p01-045— | Workflow-evidence taxonomy with court-grade provenance: WorkflowEvidence: {stated, document, recording, integration_read, public_observation} each with defined reliability ordering used by ranking | hypothesiscandidate | evidence-class discipline specialized for discovery |
| — | inn-p01-046— | Dead-end insurance: every discovery flow keeps a 'what we'd do with zero further answers' fallback spec so a client who ghosts mid-discovery still yields a usable (widely-hedged) proposal | hypothesiscandidate | lane reasoning on client-effort reality |
| — | inn-p01-047— | Cross-industry analog suggestion: when within-industry supply is thin, discovery may propose an analog from a sibling archetype ('what mortgage brokers do for conditions works for your change-orders') with the analogy recorded as inference | hypothesiscandidate | archetype compression (A02) exploited at discovery time |
| — | inn-p01-048— | Zero-question fast lane for tiny scopes: below a size threshold (one workflow, one system), skip structured discovery and go straight to prototype-as-question (P02 seam), measuring whether skipping hurts acceptance | hypothesiscandidate | P02 open question 'when should a prototype answer a question' |
| — | inn-p01-049— | Client-language mirror check: before handoff, render the captured ClientContext back in the client's own vocabulary for sign-off; signed context becomes the P02 input baseline and disputes become recorded contradictions | hypothesiscandidate | read-back verification pattern from atom contracts applied to discovery |
| — | inn-p01-050— | Regulatable evidence budget: discovery records exactly which external sources were touched per engagement (scrape log with robots/consent state) making the 'what may be scraped automatically' question an auditable policy rather than an open debate | hypothesiscandidate | parts.json P01 open question on scraping boundary |
Complete source register
115 mined records. Search and inspect the full agent output; nothing is collapsed into a top-ten headline.
| Rank | Identity | What the agent found | Evidence | Source |
|---|---|---|---|---|
| — | Auctorai-discovery | AI agents join enterprise software implementation discovery and refinement calls, capture requirements as they surface, and generate SOWs, proposals, architecture docs, user stories, test plans and configurations from them. | Drelevant | Source ↗ |
| — | Cloobot Xai-discovery | A 'CRM for Requirements' that converts workshops, stakeholder conversations, notes and SoWs into structured requirements with live traceability from requirement to user story to Jira ticket to release. | Drelevant | Source ↗ |
| — | Fin for Sales (Intercom)ai-discovery | An inbound AI SDR that runs sales conversations end to end — engaging visitors, guiding discovery on pricing/features/objections, qualifying against a playbook, and booking a meeting, starting a trial, or handing off with full context. | Drelevant | Source ↗ |
| — | Avomaconversation-intelligence | All-in-one AI meeting assistant and conversation/revenue intelligence platform that records, transcribes, and structures meetings, and automatically maps meeting content into sales-methodology CRM fields. | Drelevant | Source ↗ |
| — | Gongconversation-intelligence | Revenue AI platform that automatically captures every customer interaction, maps it to business context in a 'Revenue Graph', and surfaces deal risk and opportunity insights. | Drelevant | Source ↗ |
| — | Lawmaticsintake | AI-powered legal CRM whose intake product turns inquiries into booked consults and signed agreements, with automatic conflict scanning and AI lead scoring. | Drelevant | Source ↗ |
| — | Loopiointake | RFP, DDQ and security-questionnaire response platform that auto-populates answers from a curated content Library and scores every AI-generated answer for confidence. | Drelevant | Source ↗ |
| — | Conveyorintake | Security-questionnaire automation that auto-completes complex questionnaires by pulling from existing documents, wikis and external sites rather than requiring a manually curated answer library. | Irelevant | Source ↗ |
| — | Zoovuguided-selling | AI product-discovery platform unifying search, guided-selling assistants, product finders, quizzes and CPQ/configuration on one engine for B2C and B2B ecommerce. | Drelevant | Source ↗ |
| — | excentosguided-selling | Guided-selling software (Product Finder, Product Configurator, Set Builder, Troubleshooting Guide) blending traditional guided selling with generative AI to act as a digital advisor. | Drelevant | Source ↗ |
| — | Ignitionintake | Proposal-to-payment platform for professional services: online proposals that double as engagement agreements, capturing payment details at signature and automating billing thereafter. | Drelevant | Source ↗ |
| — | Karbonintake | Accounting practice-management platform combining workflow automation, email triage, client collaboration and an AI coworker (Kai), with onboarding delivered as templated workflows and client requests. | Dadjacent | Source ↗ |
| — | Demandbaseenrichment | Account intelligence platform that identifies anonymous website visitors as companies, enriches CRM records at scale, and combines firmographic, technographic, buyer and intent data into an account view. | Drelevant | Source ↗ |
| — | HubSpot Breeze Intelligence (formerly Clearbit)enrichment | Native HubSpot enrichment layer, the absorbed remains of Clearbit, offering firmographic enrichment and smart form shortening inside the CRM. | Irelevant | Source ↗ |
| — | Clayenrichment | GTM data platform running waterfall enrichment across 100+ data providers, with an AI research agent (Claygent) that browses the web to answer custom research questions per account. | Irelevant | Source ↗ |
| — | Apollo.ioenrichment | Self-serve B2B database and sales-engagement platform with an enrichment API, claiming 230M+ contacts and 30M+ companies. | Iadjacent | Source ↗ |
| — | ZoomInfoenrichment | Largest B2B database (claimed 500M+ contacts) with a company enrichment API returning firmographics, revenue, headcount, location, industry codes and social presence from a domain, name or ZoomInfo ID in one call. | Iadjacent | Source ↗ |
| — | Chorus by ZoomInfoconversation-intelligence | Conversation intelligence enriched with ZoomInfo buyer intent and firmographic data, adding context about who is on the call directly into call analysis. | Iadjacent | Source ↗ |
| — | Clari (incl. Clari Copilot / Wingman)conversation-intelligence | Revenue platform centred on forecasting and pipeline inspection, with Clari Copilot (formerly Wingman) providing conversation intelligence and real-time call guidance. | Iadjacent | Source ↗ |
| — | Salesloftconversation-intelligence | Revenue orchestration platform with sales engagement, deal management, forecasting and conversation intelligence. | Iweak | Source ↗ |
| — | Fireflies.aiconversation-intelligence | Meeting recorder and AI notetaker with the broadest platform compatibility (Zoom, Teams, Google Meet, RingCentral, Webex, uploaded audio) and transparent tiered pricing including a free plan. | Iadjacent | Source ↗ |
| — | Jiminnyconversation-intelligence | Conversation intelligence and coaching platform positioned for mid-market revenue teams. | Iadjacent | Source ↗ |
| — | Revenue.ioconversation-intelligence | Salesforce-centric conversation intelligence with real-time coaching during live calls and automated methodology scoring. | Iadjacent | Source ↗ |
| — | Clio Growintake | Legal practice management suite with client intake forms, scheduling and onboarding integrated into matter management. | Irelevant | Source ↗ |
| — | PracticePantherintake | Legal practice management with customisable intake forms, e-signatures and automated client onboarding. | Iadjacent | Source ↗ |
| — | Leglintake | UK legal client onboarding and lifecycle platform combining automated onboarding with built-in KYC, KYB and AML compliance checks. | Irelevant | Source ↗ |
| — | Content Snareintake | Client information and document collection tool built specifically for the problem of chasing clients for content, assets and answers. | Irelevant | Source ↗ |
| — | Dubsadointake | Client management platform for freelancers and small agencies automating the full intake sequence: forms, proposals, contracts, invoicing and workflows. | Irelevant | Source ↗ |
| — | HoneyBookintake | All-in-one clientflow platform for independent service businesses covering inquiry capture, proposals, contracts, payments and scheduling. | Iweak | Source ↗ |
| — | Bonsaiintake | Freelancer/agency business platform bundling proposals, attorney-vetted contract templates, invoicing, accounting and tax tools. | Iweak | Source ↗ |
| — | Financial Centsintake | Accounting practice management platform with client onboarding via workflow templates and automated client document chasing. | Dadjacent | Source ↗ |
| — | TaxDomeintake | Accounting practice management with client portal, organisers, e-signature and pipeline-based onboarding. | Irelevant | Source ↗ |
| — | Canopyintake | Accounting practice management suite with client portal, document management and engagement workflows. | Iweak | Source ↗ |
| — | Moxointake | Client interaction and workflow automation platform used for onboarding in legal, financial services and professional services, coordinating tasks and document exchange in a shared workspace. | Iadjacent | Source ↗ |
| — | Motion.iointake | Client onboarding and collaboration platform for agencies and firms, automating the client journey through portals and workflows. | Iadjacent | Source ↗ |
| — | Assemblyintake | White-label client portal software serving law firms, accounting and bookkeeping practices, marketing agencies and consultancies, with secure file sharing and automations. | Iweak | Source ↗ |
| — | Osprey Approachintake | UK legal practice suite combining case management, practice management, legal accounting and client onboarding with digital identity verification. | Iweak | Source ↗ |
| — | Clientmanager.iointake | Client onboarding and management tool aimed at agencies, covering onboarding checklists and client tracking. | Iweak | Source ↗ |
| — | Plutiointake | All-in-one business management tool for freelancers covering projects, proposals, invoices and client onboarding. | Iweak | Source ↗ |
| — | Responsive (formerly RFPIO)intake | Response management platform for high-volume RFPs and security questionnaires, with AI-assisted drafting from a centralised content library and workflow governance across concurrent projects. | Iadjacent | Source ↗ |
| — | Vantaintake | Compliance automation platform that bundles a trust centre and security questionnaire support alongside continuous control monitoring. | Iadjacent | Source ↗ |
| — | Drataintake | Compliance automation with continuous control monitoring and trust centre, competing directly with Vanta. | Iweak | Source ↗ |
| — | SafeBaseintake | Trust-centre-first platform that pre-empts security questionnaires by publishing a self-serve security posture page. | Iadjacent | Source ↗ |
| — | Whisticintake | Vendor security assessment and trust-profile platform for both sending and answering security assessments. | Iweak | Source ↗ |
| — | Outreach (AI Prospecting Agent)ai-discovery | Sales execution platform whose AI Prospecting Agent researches accounts, generates messaging, prioritises targets and enrols prospects into sequences. | Iadjacent | Source ↗ |
| — | Persana AIai-discovery | Autonomous AI sales agent managing outreach from lead discovery to meeting booking, finding qualified leads and personalising with live data. | Iweak | Source ↗ |
| — | SellScale / Selixai-discovery | AI sales agent (Selix) that posts updates in Slack explaining what it is researching and what it discovered, and proposes actions for human review before execution. | Drelevant | Source ↗ |
| — | MindStudioai-discovery | AI agent building platform; publishes an explicit discovery-question methodology for scoping and pricing AI automation engagements, and ships a product-manager agent ('Remy'). | Dadjacent | Source ↗ |
| — | Inleadoenrichment | Purpose-built inbound demo-request enrichment: prospect supplies name and business email, the platform extracts the domain, appends revenue, employees, industry and location in milliseconds, then routes by fit. | Irelevant | Source ↗ |
| — | Exploriumenrichment | External data platform with a documented Firmographics enrichment API endpoint returning revenue, company size, industry classification and other attributes. | Dadjacent | Source ↗ |
| — | Cognismenrichment | European-focused B2B data platform with GDPR-structural compliance, DNC screening across 12 EU countries, and phone-verified contact data. | Iweak | Source ↗ |
| — | CanvasLogicguided-selling | CPQ combining configuration logic with 3D visual product configuration for modular, variant-rich manufactured products. | Iadjacent | Source ↗ |
| — | PROS Smart CPQguided-selling | CPQ with an AI pricing and recommendation engine that analyses past sales records and market data to identify optimal pricing, upsell and bundling. | Iadjacent | Source ↗ |
| — | SAP CPQguided-selling | Enterprise CPQ supporting user-guided sales processes, multi-level approvals and advanced pricing structures for complex assembly configuration. | Iweak | Source ↗ |
| — | Outfindoguided-selling | Buyer-facing product-selection guide for ecommerce and self-serve B2B, converting product complexity into a short question flow. | Iadjacent | Source ↗ |
| — | process-intelligence-solutions/pm4pyprocess-mining | Reference Python process-mining library: discovery (Alpha/Heuristics/Inductive miner), conformance checking, XES + OCEL 2.0 event logs, Petri nets/BPMN/process trees. | Drelevant | Source ↗ |
| — | apromore/ApromoreCoreprocess-mining | Open-source Core Edition of the Apromore process-mining platform: log/BPMN importer, portal, Process Discoverer, business calendar, model editor. | Dweak | Source ↗ |
| — | bupaverse/bupaRprocess-mining | Core R package of the bupaR ecosystem for business process analysis; event-log object model, case/trace statistics, companion packages for discovery and performance analysis. | Drelevant | Source ↗ |
| — | bupaverse/pm4pyprocess-mining | R interface bridging bupaR to the Python PM4Py library; exposes inductive miner discovery, Petri net rendering and conformance alignments to R users. | Iadjacent | Source ↗ |
| — | computertechworld/ProcessMiningprocess-mining | Curated list of open-source process-mining tools and resources (ProM, Apromore CE, RapidProM, OpenXES). | Iadjacent | Source ↗ |
| — | TheWoops/awesome-processminingprocess-mining | Collection of process-mining resources. | Iweak | Source ↗ |
| — | RPA-US/screenrpatask-mining | Screen-based task-mining framework for RPA: Django platform capturing UI logs, eye-tracking gaze logs and screenshots, with modules for behaviour monitoring, decision discovery, feature extraction, process discovery and reporting. | Dweak | Source ↗ |
| — | apromore/RPA_UILoggertask-mining | User-interaction logger for robotic process mining; Excel add-in passive listener plus Chrome extension, writing captured UI events to CSV. | Dadjacent | Source ↗ |
| — | OpenAdaptAI/OpenAdapttask-mining | Compiles a demonstrated GUI workflow into a deterministic, locally executable program; captures structural, accessibility, visual, OCR, spatial and transition evidence across browser, Windows, macOS, Linux, RDP and Citrix/VDI. | Drelevant | Source ↗ |
| — | xlang-ai/OpenCUAtask-mining | NeurIPS 2025 spotlight framework for computer-use agents: AgentNetTool cross-platform recorder, AgentNet dataset of 22.6K human-annotated desktop tasks across 200+ apps, AgentNetBench evaluator, OpenCUA 7B/32B/72B models. | Drelevant | Source ↗ |
| — | ActivityWatch/activitywatchtask-mining | Cross-platform automated time tracker; aw-watcher-window logs focused application and window title, aw-watcher-afk infers active/idle from keyboard and mouse, aw-watcher-web browser extension tracks active tab title and URL. | Drelevant | Source ↗ |
| — | humblebanana/open-record-replaytask-mining | macOS record-and-replay workflow recorder for computer-use agents; captures mouse, keyboard and UI events as structured traces (session.json, events.jsonl), validates recording quality and packages evidence for an agent skill-creation flow. | Iadjacent | Source ↗ |
| — | A9T9/RPAtask-mining | Ui.Vision open-source RPA with computer vision, OCR, Anthropic Computer Use/LLM integration and a web macro recorder; Selenium IDE import/export. | Iweak | Source ↗ |
| — | aisingapore/TagUItask-mining | Free RPA tool by AI Singapore covering websites, desktop apps and CLI; cross-platform; flows writable in 22 languages. | Iweak | Source ↗ |
| — | robocorp/rpaframeworktask-mining | Collection of open-source RPA libraries for Robot Framework and Python. | Iweak | Source ↗ |
| — | firecrawl/fire-enrichenrichment | Multi-agent enrichment pipeline turning an email/domain into a company profile: industry, business model, employee count, founding year, HQ, funding stage and investors, tech stack, plus custom fields — every field carrying source URLs. | Drelevant | Source ↗ |
| — | Lissy93/web-checkenrichment | All-in-one website OSINT tool running ~40 checks: DNS, SSL/TLS config and cipher suites, HTTP and security headers, cookies, Whois, robots.txt, sitemap/listed pages, open ports, traceroute, trackers, tech stack via Wappalyzer fingerprints, email config (DMARC/DKIM/SPF/BIMI), WAF detection, Wayback archive history, Tranco global rank, screenshots. | Drelevant | Source ↗ |
| — | specfy/stack-analyserenrichment | Extracts 700+ technologies from a repository — languages, SaaS products, cloud, infrastructure, dependencies and services — by parsing package.json, docker-compose.yml, go.mod and similar manifests, and mapping folder-to-service relationships. | Drelevant | Source ↗ |
| — | mattvinall/Quick-Enrich-Toolsenrichment | Open-source suite of six lead-enrichment tools intended to be cloned, customized and self-hosted with your own API keys; reference stack Railway + Vercel + Supabase + Resend at roughly $10-20/mo. | Iweak | Source ↗ |
| — | eracle/OpenOutreachenrichment | Self-hosted CLI agent for B2B lead generation: describe a product, it finds matching people from a licensed data provider and explains the fit, exporting CSV. Claims zero platform-ToS surface — browserless, no social account, no scraping. | Iadjacent | Source ↗ |
| — | taranis-ai/taranis-aienrichment | AI-assisted OSINT tool for information gathering and situational analysis. | Iweak | Source ↗ |
| — | jivoi/awesome-osintenrichment | Curated list of OSINT tools and resources. | Iweak | Source ↗ |
| — | openclarity/apiclarityintrospection | Captures and analyzes API traffic to reconstruct OpenAPI specifications with no code instrumentation, then diffs observed traffic against approved specs to flag Shadow APIs (observed but undocumented) and Zombie APIs (deprecated but still live). | Dadjacent | Source ↗ |
| — | grokify/traffic2openapiintrospection | Normalizes HTTP traffic from HAR files, Postman collections, Playwright/Puppeteer/Cypress captures, mitmproxy/Charles logs and Go http.Client into a versioned intermediate representation ('ir.v1'), then infers OpenAPI 3.0/3.1/3.2 specs — detecting path/query params, types and formats, auth schemes and pagination patterns, and diffing specs for breaking changes. | Drelevant | Source ↗ |
| — | doitintl/DoiT-AdminPulse-for-Workspaceintrospection | Google Sheets add-on for Workspace admins: security-posture checklist plus read-only Admin SDK inventory scripts covering users, license assignments, Groups settings and membership, mobile devices, shared drives, OAuth tokens (getOAuthTokens.gs), app passwords, OUs, domains and DNS. | Drelevant | Source ↗ |
| — | psignoret/azure-ad-permissions-gistintrospection | Widely-used community PowerShell script enumerating all delegated and application permission grants in Entra ID, iterating app permissions directly rather than parsing logs; updated to handle tenants with 1000+ grants. | Iadjacent | Source ↗ |
| — | walidamro-msft/entra-id-graph-consent-toolkitintrospection | Entra ID governance toolkit to list, audit and bulk-delete Microsoft Graph delegated user consents, forcing least-privilege re-consent; PowerShell 5.1 and 7+, revocation path uses pure REST with no module dependency. | Iadjacent | Source ↗ |
| — | MianAzeem/api-snifferintrospection | Playwright-based tool capturing real-time API traffic from web applications and generating structured API intelligence — schema inference, automatic endpoint classification (AUTH, USER), and an OpenAPI generator. | Iadjacent | Source ↗ |
| — | NangoHQ/nangointrospection | Self-hostable product-integration platform providing managed OAuth and unified syncs across a large catalog of third-party SaaS APIs. | Iadjacent | Source ↗ |
| — | akto-api-security/aktointrospection | Open-source API security platform whose code-analysis tool scans source repositories to discover APIs including shadow APIs, integrating with traffic-based inventory and CI/CD to catch new APIs in pull requests. | Iadjacent | Source ↗ |
| — | opticdev/opticintrospection | Captures real HTTP traffic from production or elsewhere and creates OpenAPI from it, for documentation, mocks, SDKs or contract testing. | Iadjacent | Source ↗ |
| — | schemathesis/schemathesisintrospection | Property-based API testing driven by OpenAPI and GraphQL schemas. | Iweak | Source ↗ |
| — | osquery/osqueryintrospection | Exposes an operating system as a relational database, allowing SQL queries over installed applications, running processes, browser extensions and network connections. | Iadjacent | Source ↗ |
| — | zeek/zeekintrospection | Network security monitor generating structured logs of application traffic; can reveal SaaS usage by inspecting DNS queries, IP connections and TLS/SNI metadata without decrypting content. | Iadjacent | Source ↗ |
| — | modAL-python/modALactive-learning | Modular active-learning framework on scikit-learn: ActiveLearner with pluggable query strategies (uncertainty, entropy, custom), Committee/CommitteeRegressor for query-by-committee, BayesianOptimizer, ranked batch-mode sampling and information density; Keras integration. | Drelevant | Source ↗ |
| — | scikit-activeml/scikit-activemlactive-learning | Comprehensive active-learning library on scikit-learn implementing 60+ query strategies drawn from 40+ research papers, spanning pool-based and stream-based settings, with deep active learning via skorch/PyTorch. | Drelevant | Source ↗ |
| — | usnistgov/optbayesexptactive-learning | Optimal Bayesian Experimental Design for choosing measurement settings when a parametric model is known; selects settings likely to make parameter estimates more precise, aimed at cases where measurements are expensive. | Irelevant | Source ↗ |
| — | experimental-design/bofireactive-learning | Framework for experimental design and multi-objective Bayesian optimization, built on Meta's BoTorch. | Iadjacent | Source ↗ |
| — | emdgroup/baybeactive-learning | General-purpose toolbox for Bayesian Design of Experiments focused on real-world experimental campaigns; claimed use in industry and academia. | Iadjacent | Source ↗ |
| — | VlachosGroup/nextorchactive-learning | Experimental design and Bayesian optimization library in Python/PyTorch. | Iweak | Source ↗ |
| — | mhinne/BNQDactive-learning | Toolbox for Bayesian Nonparametric Quasi-Experimental Design in Python. | Iweak | Source ↗ |
| — | bayesflow-org/bayesflowactive-learning | Amortized Bayesian inference library; ships a Bayesian Experimental Design example notebook framing BED as combining information theory and Bayesian modelling to plan experiments maximizing information gain. | Iadjacent | Source ↗ |
| — | microsoft/graphragorg-graph | Modular graph-based RAG system; an LLM-driven pipeline that extracts structured entities and relationships from unstructured narrative text and builds knowledge-graph memory structures used as targeted context for question answering. | Drelevant | Source ↗ |
| — | getzep/graphitiorg-graph | Framework for real-time temporal knowledge graphs for AI agents: incremental ingestion without full recomputation, bi-temporal fact validity windows with automatic invalidation on contradiction, provenance via 'episodes' linking every fact to its source, hybrid semantic + BM25 + graph-traversal retrieval, custom ontology through Pydantic entity and edge types. | Drelevant | Source ↗ |
| — | zjunlp/DeepKEorg-graph | Open toolkit for knowledge-graph extraction and construction (EMNLP 2022): named entity recognition, relation extraction, attribute extraction, event extraction and relational triple extraction, across standard, few-shot, document-level and multimodal settings; DeepKE-LLM branch, OneKE multi-agent extraction framework, MCP service tools. | Drelevant | Source ↗ |
| — | robert-mcdermott/ai-knowledge-graphorg-graph | Takes unstructured text, uses any OpenAI-compatible LLM (Ollama, LM Studio, vLLM, LiteLLM, OpenAI) to extract subject-predicate-object triplets, standardizes entities across document chunks, infers relationships to connect disconnected subgraphs, and renders an interactive graph. | Iadjacent | Source ↗ |
| — | junhewk/simple-graph-builderorg-graph | Obsidian plugin building a lightweight knowledge graph from notes via LLM entity extraction over a fixed 10-type ontology (PERSON, ORGANIZATION, CONCEPT, PROJECT, TOOL, EVENT, PLACE, DOCUMENT, METHOD, TOPIC) with free-form relationship verbs; hybrid entity resolution resolving most entities by hash lookup and reserving embedding search and LLM calls for genuinely ambiguous cases. | Iadjacent | Source ↗ |
| — | MadsDoodle/Knowledge-Graph-using-RDF-LLMorg-graph | Pipeline extracting entities and relationships from PDF, TXT and DOCX into RDF graphs, combining SpaCy rule-based NER with LLM extraction, plus deduplication and merging; SPARQL-queryable. | Iweak | Source ↗ |
| — | dylanhogg/llmgraphorg-graph | Creates knowledge graphs in GraphML, GEXF and HTML from a source entity's Wikipedia page, with customised prompts per entity type and cache support for iteratively growing a graph. | Iweak | Source ↗ |
| — | FabioYanezRomero/Knowledge-Graph-Builderorg-graph | Builds knowledge graphs via generative models through Ollama/Gemini/LM Studio, with structured triple extraction source-grounded via langextract, graph augmentation to bridge disconnected components, schema-constrained customizable domains, Cytoscape.js visualization and GraphML output. | Iadjacent | Source ↗ |
| — | industry-atom-specifications.mdlocal-estate | 17 industry bounded specs, 66 use cases, 12 atoms, 170-cell readiness trace; all cells readiness_unexecuted | Erelevant | research/actionmodel-builder-research-2026-08-26/phase-2/outputs/industry-atom-specifications.md |
| — | niche-atom-block-join.mdlocal-estate | 12-atom contract schema, use-case→atom map, archetype priority scoring, preserved catalogue contradictions | Erelevant | research/actionmodel-builder-research-2026-08-26/expansion/outputs/niche-atom-block-join.md |
| — | actionist-solutions-sweep-spec-2026-08-26.mdlocal-estate | solution_atom discovery record; builder decomposition items 1-2 are P01/P02 restated | Erelevant | research/actionist-solutions-sweep-spec-2026-08-26.md |
| — | autosaas-method.mdlocal-estate | opportunity-intake gate and spec-constitution field list; procedural design, never measured | Erelevant | [private local source withheld] |
| — | 00-MASTER-SYNTHESIS.mdlocal-estate | nine irreducible jobs; jobs 1-2 are this lane; demand compression thesis | Erelevant | knowledge/00-MASTER-SYNTHESIS.md |
| — | 02-ASSUMPTION-LEDGER.mdlocal-estate | A01/A02/A17/A22/A38 bound this lane's claims | Erelevant | knowledge/02-ASSUMPTION-LEDGER.md |
| — | parts.jsonlocal-estate | P01/P02 theses, owned outputs and open questions | Erelevant | site/system-map/data/parts.json |
| — | first-principles-framework.mdlocal-estate | C01/C02 atomic capabilities = demand normalization and requirement elicitation | Erelevant | research/actionmodel-builder-research-2026-08-26/outputs/first-principles-framework.md |
| — | synthetic-pilot-specification.mdlocal-estate | pre-registered fixture/state-machine discipline reused for acceptance design | Erelevant | research/actionmodel-builder-research-2026-08-26/phase-3/outputs/synthetic-pilot-specification.md |
| — | public-signals-wave-11.mdlocal-estate | public signal classes A/B/C and their explicit limits | Erelevant | research/actionmodel-builder-research-2026-08-26/expansion/wave-11/outputs/public-signals-wave-11.md |
Source: site/system-map/data/parts.json#P01 · preserved research datasets remain inline below their source-backed synthesis.