royal-app
by
Satish Kumar
Satish Kumarcreate this app
Comments (0)
Sign in to leave a comment
Project Tasks
112 tasks#1
Generate system requirement document
3m 46s0.2 cr used
Done#2
Generate personas & user flows
0m 9s0.2 cr used
Done#7
Create flow for Account Owner
0m 8sCredits in parent
Done#8
/
4m 12sCredits in subtasks
Done#22
Fix / review findings: Measured horizontal overflow at mobile: the page is 418px… (+2 more)
1m 36s0.4 cr used
Done#15
/ / Landing Entry Poster
0m 36s1.6 cr used
Done#16
/ / Landing Limits Ledger
0m 25s1.6 cr used
Done#17
/ / Landing Request Lifecycle
0m 30s1.6 cr used
Done#18
/ / Landing Grounding Policy
0m 23s1.6 cr used
Done#19
/ / Landing Sign In Close
0m 22s1.6 cr used
Done#20
/ / Footer
0m 16s1.6 cr used
Done#21
Footer
0m 16sCredits in parent
Done#9
Assistant
3m 31sCredits in subtasks
Done#34
Fix Assistant review findings: Q1 as reviewed now: Footer is placed alongside the top of… (+3 more)
1m 0s0.4 cr used
Done#23
Assistant / Navigation
0m 16s1.6 cr used
Done#24
Assistant / Status Rail
0m 25s1.6 cr used
Done#25
Assistant / Transcript Ledger
0m 48s1.6 cr used
Done#26
Assistant / Composer Strip
0m 23s1.6 cr used
Done#27
Assistant / Footer Slot
0m 0s1.6 cr used
Done#10
Voice
6m 3sCredits in subtasks
Done#58
Fix Voice review findings: frontend/src/components/Voice Limits Strip.jsx states the… (+1 more)
0m 14s0.4 cr used
Done#29
Voice / Navigation Slot
0m 2s1.6 cr used
Done#30
Voice / Capture Console
0m 47s1.6 cr used
Done#31
Voice / Usage And Consent Ledger
0m 36s1.6 cr used
Done#32
Voice / Limits Strip
0m 33s1.6 cr used
Done#33
Voice / Footer Slot
0m 0s1.6 cr used
Done#11
Hands-Free
8m 59sCredits in subtasks
Done#60
Fix Hands-Free review findings: Hands Free Listening Console.jsx reads hands-free consent… (+1 more)
0m 55s0.4 cr used
Done#35
Hands-Free / Hands Free Navigation Slot
0m 1s1.6 cr used
Done#36
Hands-Free / Hands Free Consent Gate
0m 27s1.6 cr used
Done#37
Hands-Free / Hands Free Listening Console
0m 51s1.6 cr used
Done#38
Hands-Free / Hands Free Wake Phrase Reference
0m 37s1.6 cr used
Done#39
Hands-Free / Hands Free Limits Strip
0m 33s1.6 cr used
Done#40
Hands-Free / Hands Free Footer Slot
0m 1s1.6 cr used
Done#12
Settings
7m 51sCredits in subtasks
Done#41
Settings / Navigation Slot
0m 0s1.6 cr used
Done#42
Settings / Header
0m 21s1.6 cr used
Done#43
Settings / Profile Form
0m 44s1.6 cr used
Done#44
Settings / Resume Block
1m 48s1.6 cr used
Done#45
Settings / Limits Strip
0m 54s1.6 cr used
Done#46
Settings / Footer Slot
0m 0s1.6 cr used
Done#13
Wake Phrases
9m 17sCredits in subtasks
Done#61
Fix Wake Phrases review findings: Detection phrase names collapse into vertical letter… (+1 more)
0m 14s0.4 cr used
Done#47
Wake Phrases / Navigation Slot
0m 0s1.6 cr used
Done#48
Wake Phrases / Header
0m 24s1.6 cr used
Done#49
Wake Phrases / Editor
0m 55s1.6 cr used
Done#50
Wake Phrases / Detection Reference
0m 37s1.6 cr used
Done#51
Wake Phrases / Limits Strip
0m 52s1.6 cr used
Done#52
Wake Phrases / Footer Slot
0m 0s1.6 cr used
Done#14
Not Found
7m 34sCredits in subtasks
Done#59
Fix Not Found review findings: Footer content is visibly compressed and its links run… (+1 more)
0m 22s0.4 cr used
Done#54
Not Found / Navigation Slot
0m 0s1.6 cr used
Done#55
Not Found / Statement
0m 21s1.6 cr used
Done#56
Not Found / Route Index
0m 50s1.6 cr used
Done#57
Not Found / Footer Slot
0m 0s1.6 cr used
Done#28
Navigation
0m 16sCredits in parent
Done#5
Architecture
1m 39s0.2 cr used
Done#62
Create backend_modules diagram
1m 22sCredits in parent
Done#63
Create frontend_modules diagram
1m 22sCredits in parent
Done#64
Create sequence diagram
1m 22sCredits in parent
Done#65
Create service_network diagram
1m 22sCredits in parent
Done#6
Workspace task plan
3m 54s0.2 cr used
Done#53
Update /
Credits unavailable
Backlog#66
Assistant request control, cancellation, and workspace theming
120 cr estimated
To Do#67
Sign in to Echo Assist via OIDC authorization-code flow
100 cr estimated
To Do#68
Live-answer grounding, non-fabrication, TTS, wake phrases, and request safeguards
140 cr estimated
To Do#69
Wake-phrase settings and voice consent configuration
80 cr estimated
To Do#70
Voice reply bounds, file cleanup, usage records, and transcription
140 cr estimated
To Do#71
Bounded file storage, cautious answers, wake-phrase storage, and replay protection
120 cr estimated
To Do#72
Require signed-in identity and complete OIDC callback
100 cr estimated
To Do#73
Private assistant profile, resume upload validation, and wake-phrase authorization
120 cr estimated
To Do#74
Provide chat, transcription, and speech synthesis through the AI gateway
120 cr estimated
To Do#75
Clean up stale voice recordings and record consent per account
100 cr estimated
To Do#76
Enforce request control, usage limits, and grounding rules for assistant requests
140 cr estimated
To Do#77
Deliver the assistant workspace: navigation, composer, live answers, and spoken replies
140 cr estimated
To Do#78
Admit voice requests, bound transcripts, and store resume grounding safely
140 cr estimated
To Do#79
Manage assistant profile, resume upload, and wake-phrase defaults
120 cr estimated
To Do#80
Finalize voice and chat requests with account-scoped lifecycle handling
140 cr estimated
To Do#81
Run hands-free listening with wake-phrase detection and grounded live answers
140 cr estimated
To Do#82
Bound assistant replies and verify resume submissions
120 cr estimated
To Do#83
Admit assistant requests and guard profile writes
140 cr estimated
To Do#84
Save wake phrases and manage resume replacement and removal
100 cr estimated
To Do#85
Issue single-use resume upload tickets and validate extracted text
120 cr estimated
To Do#86
Sign in through hosted identity and provision the account record
120 cr estimated
To Do#87
Reserve voice requests and draft first-person live answers
140 cr estimated
To Do#88
Assemble live prompts and enforce non-fabrication in answers
140 cr estimated
To Do#89
Wire workspace status, welcome, and conversation controls
120 cr estimated
To Do#90
Account-scoped wake phrases, response style, and resume upload lifecycle
120 cr estimated
To Do#91
Profile settings, message validation, upload tickets, and request finalization
120 cr estimated
To Do#92
Live first-person answers with voice capture, consent, and audio meter
140 cr estimated
To Do#93
Request-ID replay protection, atomic writes, grounding precedence, and TTS reservation
140 cr estimated
To Do#94
Typed and live assistant requests with grounding, concurrency, and cancellation
140 cr estimated
To Do#95
Voice request throttling, recording lifecycle, daily cap, and live-mode capture
140 cr estimated
To Do#96
Assistant profile settings with revision conflict detection
100 cr estimated
To Do#97
Manage assistant profile instructions, style, and resume
80 cr estimated
To Do#98
Sign-in through the identity provider with account-scoped isolation
120 cr estimated
To Do#99
Send the system policy and validated messages to the AI gateway
120 cr estimated
To Do#100
Call the AI gateway with a 60-second timeout and no retries
100 cr estimated
To Do#101
Hold the owner's resume and short-lived voice recordings under ownership and deletion rules
120 cr estimated
To Do#102
Time out gateway calls after 60 seconds without retrying
100 cr estimated
To Do#103
Allow only one active assistant request per account
120 cr estimated
To Do#104
Verify resume ownership before deleting a file
80 cr estimated
To Do#107
Show the not-found page for unmatched routes
60 cr estimated
To Do#111
Build the /auth/callback route to finish sign-in
80 cr estimated
To Do#105
Wire the Hands-Free panel to live capture, wake phrases, and drafted answers
140 cr estimated
To Do#106
Wire Settings to live profile data and conflict handling
120 cr estimated
To Do#108
Build the Voice Workspace for recording, live answers, and speech
160 cr estimated
To Do#109
Build Assistant Profile Settings for instructions, style, resume, and wake phrases
160 cr estimated
To Do#110
Build the assistant profile surface for editing profile details
100 cr estimated
To Do#112
Build the Conversation Workspace for typed chat and live answers
160 cr estimated
To Do#113
Build the Assistant Workspace with typed and live answer modes
160 cr estimated
To Do#114
Build wake-phrase settings page for the account owner
100 cr estimated
To Do
EchoAssist
An assistant that says only what it can stand behind.
A private, account-scoped voice and text assistant for one owner. Ask a question in text or request a live answer mid-conversation, grounded in your own uploaded background and custom instructions — never in invented experience.
Sign-in is delegated to the identity provider; your account record is created automatically on first sign-in.
Request lifecycle
- 01SendYour message leaves the composer with a unique request ID.
- 02LeaseOne active request per account, held for 90 seconds.
- 03ProviderThe AI gateway is called once — 60-second timeout, zero retries.
- 04ValidateOutput is bounded and checked against the spoken-answer budget.
- 05SpeakThe answer is returned for you to read aloud.
Send → Lease → ProviderValidate → Speak
01 Limits
What the instrument will and will not do.
These are the working limits of the assistant for one account owner. They are stated before sign-in and applied on every request.
- Request lease
- 90 seconds
- One active request at a time; a stalled request cannot block the account indefinitely.
- Assistant requests
- 50 per UTC day
- Counter resets by UTC day.
- Minimum interval
- 3 seconds
- Rapid-fire submissions are throttled.
- Spoken-answer budget
- 1,100 characters
- Targeting under 1,000 so the answer can be said in a few breaths.
- Message history
- 20 messages
- 6,000 characters per message, 24,000 characters total.
- Voice requests
- 1,000 per UTC day
- Each transcription clip and each spoken reply counts.
6stated limits
02Lifecycle
Every request moves through five stated stops.
Request lifecyclesend → lease → provider → validate → speak
- 01Send
Your message leaves the composer with a unique request ID.
- 02Lease
One active request per account, held for 90 seconds.
- 03Provider
The AI gateway is called once — 60-second timeout, zero retries.
- 04Validate
Output is bounded and checked against the spoken-answer budget.
- 05Speak
The answer is returned for you to read aloud.
03 Grounding
Every answer stands on something you supplied.
The assistant is allowed to be useful precisely because it is bounded. These are the sources it may draw on, and the things it will never do on your behalf.
What grounds an answer
Two sources, supplied as data rather than as instructions.
- Your uploaded resume textRésumé content you provide is supplied to the assistant as a JSON data packet, never as system instructions, so document text cannot act as executable commands.Data packet, not instructions
- Your explicitly stated custom instructionsInstructions you set yourself ground the answer only where they do not conflict with your résumé facts. Résumé facts take precedence.Non-conflicting only
- Résumé facts take precedenceWhere a custom instruction conflicts with a verified résumé fact, the résumé fact wins.Fact over preference
- Style requests override presentation, never facts.An explicit style request in your custom instructions can change tone or length, but it can never loosen the factual grounding rules.Presentation only
What the assistant will never do
Hard limits, stated plainly rather than discovered later.
- Never invent or embellish.No fabricated experience, skills, tools, projects, employers, education, years, responsibilities or achievements.Non-fabrication
- Other speakers’ claims are not evidence.Something another person says during the conversation is never treated as evidence of your background.Not evidence
- Earlier drafts are never carried forward as fact.A previously generated draft is labelled non-factual and is not reused as grounded source material.Non-factual
- No live web access, no tools, no invented citations.The assistant cannot browse, cannot run tools, and never produces a citation it cannot stand behind.No tools, no browsing
- Requests to pretend or exaggerate are not evidence.Asking the assistant to invent, exaggerate or adopt unsupported credentials does not make those details true.Unsupported = excluded
Profile — grounded source blockPlain text, no avatar
RésuméNot loaded — no profile document attached
Custom instructionsNot set — default response style in use
Grounding statusGeneral knowledge only
With no profile attached, the assistant still helps with general knowledge and uses conditional language instead of making claims about your career.
EchoAssist — Echo
One account, one assistant, nothing kept.
EchoAssist is private to a single signed-in account owner, and your conversation content is never persisted — the assistant drafts and returns, and nothing of the conversation is stored.
Sign-inSign-in is delegated to the identity provider; your account record is created automatically on first sign-in, so there is no separate registration step.
- Authentication
- Delegated to the identity provider (OIDC).
- Account record
- Created automatically on first sign-in.
- Registration
- No separate registration step.
- Conversation content
- Never persisted.
EchoAssist
An assistant that says only what it can stand behind.
A private, account-scoped voice and text assistant for one owner. Ask a question in text or request a live answer mid-conversation, grounded in your own uploaded background and custom instructions — never in invented experience.
Sign-in is delegated to the identity provider; your account record is created automatically on first sign-in.
Request lifecycle
- 01SendYour message leaves the composer with a unique request ID.
- 02LeaseOne active request per account, held for 90 seconds.
- 03ProviderThe AI gateway is called once — 60-second timeout, zero retries.
- 04ValidateOutput is bounded and checked against the spoken-answer budget.
- 05SpeakThe answer is returned for you to read aloud.
Send → Lease → ProviderValidate → Speak
01 Limits
What the instrument will and will not do.
These are the working limits of the assistant for one account owner. They are stated before sign-in and applied on every request.
- Request lease
- 90 seconds
- One active request at a time; a stalled request cannot block the account indefinitely.
- Assistant requests
- 50 per UTC day
- Counter resets by UTC day.
- Minimum interval
- 3 seconds
- Rapid-fire submissions are throttled.
- Spoken-answer budget
- 1,100 characters
- Targeting under 1,000 so the answer can be said in a few breaths.
- Message history
- 20 messages
- 6,000 characters per message, 24,000 characters total.
- Voice requests
- 1,000 per UTC day
- Each transcription clip and each spoken reply counts.
6stated limits
02Lifecycle
Every request moves through five stated stops.
Request lifecyclesend → lease → provider → validate → speak
- 01Send
Your message leaves the composer with a unique request ID.
- 02Lease
One active request per account, held for 90 seconds.
- 03Provider
The AI gateway is called once — 60-second timeout, zero retries.
- 04Validate
Output is bounded and checked against the spoken-answer budget.
- 05Speak
The answer is returned for you to read aloud.
03 Grounding
Every answer stands on something you supplied.
The assistant is allowed to be useful precisely because it is bounded. These are the sources it may draw on, and the things it will never do on your behalf.
What grounds an answer
Two sources, supplied as data rather than as instructions.
- Your uploaded resume textRésumé content you provide is supplied to the assistant as a JSON data packet, never as system instructions, so document text cannot act as executable commands.Data packet, not instructions
- Your explicitly stated custom instructionsInstructions you set yourself ground the answer only where they do not conflict with your résumé facts. Résumé facts take precedence.Non-conflicting only
- Résumé facts take precedenceWhere a custom instruction conflicts with a verified résumé fact, the résumé fact wins.Fact over preference
- Style requests override presentation, never facts.An explicit style request in your custom instructions can change tone or length, but it can never loosen the factual grounding rules.Presentation only
What the assistant will never do
Hard limits, stated plainly rather than discovered later.
- Never invent or embellish.No fabricated experience, skills, tools, projects, employers, education, years, responsibilities or achievements.Non-fabrication
- Other speakers’ claims are not evidence.Something another person says during the conversation is never treated as evidence of your background.Not evidence
- Earlier drafts are never carried forward as fact.A previously generated draft is labelled non-factual and is not reused as grounded source material.Non-factual
- No live web access, no tools, no invented citations.The assistant cannot browse, cannot run tools, and never produces a citation it cannot stand behind.No tools, no browsing
- Requests to pretend or exaggerate are not evidence.Asking the assistant to invent, exaggerate or adopt unsupported credentials does not make those details true.Unsupported = excluded
Profile — grounded source blockPlain text, no avatar
RésuméNot loaded — no profile document attached
Custom instructionsNot set — default response style in use
Grounding statusGeneral knowledge only
With no profile attached, the assistant still helps with general knowledge and uses conditional language instead of making claims about your career.
EchoAssist — Echo
One account, one assistant, nothing kept.
EchoAssist is private to a single signed-in account owner, and your conversation content is never persisted — the assistant drafts and returns, and nothing of the conversation is stored.
Sign-inSign-in is delegated to the identity provider; your account record is created automatically on first sign-in, so there is no separate registration step.
- Authentication
- Delegated to the identity provider (OIDC).
- Account record
- Created automatically on first sign-in.
- Registration
- No separate registration step.
- Conversation content
- Never persisted.
No comments yet. Be the first!