candidates

Every statement both blind runs produced (3,819), with its full trail: script check → verifier verdict and reason → adjudicator ruling and reason → whether it is in the final set. Most rejections are duplicates between the two runs.

Data license: Not yet licensed — internal review only · Data source: Ahmed ElBatanony, Innopolis University

64 rows where adjudicator = "rejected (duplicate upheld)", app_id = "F1-005" and in_final_set = "0.0"

✎ View and edit SQL

This data as json, CSV (advanced)

Suggested facets: source_url, source_file

run 2

  • b 62
  • a 2

verifier_verdict 1

in_final_set 1

app_id 1

  • DeskcommCRM · 64 ✖

adjudicator 1

  • rejected (duplicate upheld) · 64 ✖
candidate_id ▼ app_id run statement excerpt source_url source_file script_check verifier_verdict verifier_reason verifier_correction adjudicator adjudicator_reason in_final_set
F1-005-A007 DeskcommCRM F1-005 a The system automatically blocks an AI message that promises something not in the catalogue The guardrail blocked it: that lead time doesn’t exist in your catalogue. https://deskcomm.com.br/en 02.txt passed rejected duplicate of F1-005-A006 (the promise check is one of the seven checks)   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-A052 DeskcommCRM F1-005 a The system lets a user send in-app notices with severities and stuck-message alerts a notice center with severities, stuck-message alerts, send-protection controls (window/pace/cap) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A013; the notice center is system-generated, users do not send notices.   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B001 DeskcommCRM F1-005 b The system lets an AI agent answer customers on WhatsApp AI agents that answer, qualify and sell on WhatsApp https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A001   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B002 DeskcommCRM F1-005 b The system lets an AI agent qualify leads AI agents with per-tenant RAG answer customers, qualify leads, move them through the pipeline, trigger automations and know when to hand off to a human https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A003   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B003 DeskcommCRM F1-005 b The system lets an AI agent move leads through the pipeline move them through the pipeline https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A004   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B005 DeskcommCRM F1-005 b The system lets an AI agent hand off a conversation to a human know when to hand off to a human https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A008   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B006 DeskcommCRM F1-005 b The system lets an AI agent answer customers using a per-tenant knowledge base (RAG) per-tenant RAG https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A005   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B009 DeskcommCRM F1-005 b The system lets a user analyze the sentiment of conversations sentiment analysis https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A024   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B010 DeskcommCRM F1-005 b The system lets a user assign a conversation to the AI as an assignee AI as a first-class assignee https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A025   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B011 DeskcommCRM F1-005 b The system lets a manager set a spending cap per organization for AI usage per-org spending caps https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A023   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B012 DeskcommCRM F1-005 b The system automatically revives a cold conversation with follow-ups using adaptive timing and per-stage triggers follow-up that revives a cold conversation (with adaptive timing and per-stage triggers) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A010   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B013 DeskcommCRM F1-005 b The system lets a user see a radar of conversations at risk of dying unanswered a radar of what's at risk of dying unanswered https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A011   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B014 DeskcommCRM F1-005 b The system lets a user see a notice center for items that need a human decision a notice center for what needs a human decision https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A013   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B015 DeskcommCRM F1-005 b The system automatically turns resolved conversations into new knowledge for the AI agent resolved conversations become new knowledge https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A015   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B016 DeskcommCRM F1-005 b The system lets a user see whether the AI agent is improving, where it fails and what is left to teach on an AI Evolution screen the **AI Evolution** screen shows whether the agent is improving, where it fails and what's left to teach https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A014   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B017 DeskcommCRM F1-005 b The system lets a user review improvements that the AI suggests for itself and apply them as a new version **Proposals** are improvements the AI suggests for itself, applicable as a new version https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A016   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B018 DeskcommCRM F1-005 b The system lets a user configure a vocabulary per pipeline configurable vocabulary per pipeline https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A017   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B019 DeskcommCRM F1-005 b The system lets a user connect WhatsApp numbers by QR code via **QR code** (WAHA, multi-number, with anti-ban: throttle + jitter + time windows) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A018   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B020 DeskcommCRM F1-005 b The system lets a user connect WhatsApp through the official messaging channel of the platform provider or through **Meta's official channel** (Cloud API, with approved templates kept in sync) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A019   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B021 DeskcommCRM F1-005 b The system automatically throttles WhatsApp sending with jitter and time windows to avoid number bans anti-ban: throttle + jitter + time windows https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A020   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B022 DeskcommCRM F1-005 b The system automatically detects opt-out messages from a contact STOP detection https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 01.txt passed rejected duplicate of F1-005-A021   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B023 DeskcommCRM F1-005 b The system lets a user choose which AI provider the system uses and switch it later per part of the system decided at install time and switchable later from the screen, **per part of the system** https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A022   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B024 DeskcommCRM F1-005 b The system lets a manager assign and transfer conversations with an audit trail audited assignment/transfer https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A027   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B025 DeskcommCRM F1-005 b The system automatically distributes conversations to agents through a round-robin queue round-robin queue https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A028   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B026 DeskcommCRM F1-005 b The system automatically routes conversations by detected intent automatic intent routing https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A029   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B027 DeskcommCRM F1-005 b The system lets an admin limit what each role can see with per-role visibility scopes per-role visibility scopes https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A030   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B028 DeskcommCRM F1-005 b The system lets a user create capture sources that receive leads from landing pages, custom forms or external tools Every tenant can create **capture sources**: a public endpoint (`/api/v1/webhooks/in/`) that receives leads from landing pages, custom forms or tools like Zapier/n8n via POST https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A037   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B029 DeskcommCRM F1-005 b The system automatically adds a lead received through a capture source to the chosen pipeline and stage drops them straight into the chosen pipeline/stage https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A037   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B030 DeskcommCRM F1-005 b The system lets a user build WHEN/IF/THEN automation rules on CRM events tenants build **automations**: WHEN/IF/THEN rules https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A038   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B031 DeskcommCRM F1-005 b The system automatically adds tags to leads when an automation rule fires add tags https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A038   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B032 DeskcommCRM F1-005 b The system automatically moves leads between stages when an automation rule fires move leads https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A038   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B033 DeskcommCRM F1-005 b The system automatically assigns agents to leads when an automation rule fires assign agents https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A038   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B034 DeskcommCRM F1-005 b The system automatically sends WhatsApp messages when an automation rule fires send WhatsApp messages https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A038   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B035 DeskcommCRM F1-005 b The system automatically notifies an external system through an outgoing webhook when an automation rule fires notify external systems via outgoing webhooks https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A038   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B036 DeskcommCRM F1-005 b The system automatically creates automation rules paused until a user reviews and enables them Automations** (build rules, which are always born paused until reviewed and enabled) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A041   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B037 DeskcommCRM F1-005 b The system lets a manager or admin create a source, copy the ready-made endpoint, fire a test lead and see recent deliveries **Receive data** (create a source, copy the ready-made endpoint/form, fire a test lead, see recent deliveries) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A039   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B038 DeskcommCRM F1-005 b The system lets a manager or admin view a timeline of each automation run with per-action results and retry a failed external webhook call manually **Activity** (a timeline of each run, with per-action results and manual retry when an external webhook call fails) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A040   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B039 DeskcommCRM F1-005 b The system lets a user converse with customers on WhatsApp in an inbox alongside the AI **Inbox** (WhatsApp conversations, you and the AI side by side) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A002   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B040 DeskcommCRM F1-005 b The system lets a user see who went cold and is still open **Radar** (who went cold and is still open) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A011   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B041 DeskcommCRM F1-005 b The system lets a user use quick replies **Quick replies** https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A045   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B042 DeskcommCRM F1-005 b The system lets a user see where each deal sits in the funnel on a kanban board **Kanban** (where each deal sits in the funnel) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A042   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B043 DeskcommCRM F1-005 b The system lets a user manage contacts pipeline management from the screen, customer 360, contacts, tags https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A043   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B044 DeskcommCRM F1-005 b The system lets a user manage pipelines with stages, business vocabulary and loss reasons **Pipelines** (stages, business vocabulary and loss reasons) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A044   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B047 DeskcommCRM F1-005 b The system lets a user manage the knowledge base of AI agents **Knowledge** (RAG) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A051   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B053 DeskcommCRM F1-005 b The system lets a user view funnel and per-agent performance metrics **Performance** (funnel and per-agent metrics) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A046   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B054 DeskcommCRM F1-005 b The system lets a user view an audit log **Audit Log** https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 01.txt passed rejected duplicate of F1-005-A050   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B055 DeskcommCRM F1-005 b The system lets a user manage team members **Team** https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A047   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B056 DeskcommCRM F1-005 b The system lets a user configure support distribution **Support distribution** https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A028 (support distribution screen is the round-robin/routing configuration)   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B057 DeskcommCRM F1-005 b The system lets a user manage API tokens **API Tokens** https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 01.txt passed rejected duplicate of F1-005-A048   rejected (duplicate upheld) duplicate upheld 0.0
F1-005-B058 DeskcommCRM F1-005 b The system lets a user manage security settings such as MFA, recovery codes and sessions **Security** (MFA, recovery codes, sessions) https://raw.githubusercontent.com/melgarafael/DeskcommCRM/HEAD/README.en.md 04.txt passed rejected duplicate of F1-005-A049   rejected (duplicate upheld) duplicate upheld 0.0

Next page

Advanced export

JSON shape: default, array, newline-delimited, object

CSV options: 

CREATE TABLE [candidates] ([candidate_id] TEXT PRIMARY KEY, [app_id] TEXT, [run] TEXT, [statement] TEXT, [excerpt] TEXT, [source_url] TEXT, [source_file] TEXT, [script_check] TEXT, [verifier_verdict] TEXT, [verifier_reason] TEXT, [verifier_correction] TEXT, [adjudicator] TEXT, [adjudicator_reason] TEXT, [in_final_set] REAL, FOREIGN KEY ([app_id]) REFERENCES apps(app_id));
CREATE INDEX [idx_candidates_app_id] ON [candidates]([app_id]);
CREATE INDEX [idx_candidates_run] ON [candidates]([run]);
CREATE INDEX [idx_candidates_verifier_verdict] ON [candidates]([verifier_verdict]);
CREATE INDEX [idx_candidates_adjudicator] ON [candidates]([adjudicator]);