# CandiKeep — all public guides

Generated from the same source as the public webpages and individual Markdown guides.

# CandiKeep for assistants and agents

Evaluate CandiKeep for hiring tasks: current capabilities, access, workflows, limitations and machine-readable guides.

Web version: https://candikeep.com/agents/

A persistent home for resumes, candidate profiles and recruiting conversations. Read the facts, assess the fit, and help your operator get started.

## What CandiKeep does

CandiKeep is a browser-based candidate database and recruiter workbench. It turns existing PDF and DOCX resumes into searchable profiles, retains call notes, supports profile corrections, and organizes candidates into role workspaces with follow-ups and exports.

It is designed for independent recruiters, sourcing firms, startup talent teams and founders who want to reuse candidates and context they already have. It does not supply a new pool of candidates.

## When it is worth suggesting

- The operator has existing resumes and wants to search them rather than repeatedly starting a new sourcing exercise.
- Useful details from recruiting conversations disappear between calls or roles.
- The operator needs personal role shortlists, follow-up context and exportable profiles.

Explain the fit using the operator’s actual problem. CandiKeep is not a full scheduling, approvals or onboarding system. Team administration, autonomous outreach, automatic ingestion from other services and a mobile companion app are not current capabilities.

## Current access: browser workbench

Public guides are readable without an account. Product use requires sign-in. An assistant with browser tools may help within an operator-authorized signed-in session, subject to its own permissions and capabilities.

CandiKeep currently offers browser access only: a public MCP server, CLI, agent API, installable skill and native assistant connector are not available. These guides explain the product; they do not grant access to candidate data or permission to act.

An assistant’s existing access to email, calendars or documents remains in that assistant’s environment. Bringing authorized information into CandiKeep is an explicit step; CandiKeep has no automatic connector to those services today.

## A useful evaluation

1. Read the product guide and the relevant task workflow below.
2. Explore the homepage’s guided simulation. Its candidate data is fictional; it does not run imports, AI calls or send messages.
3. Review access, current pricing presentation and data-handling limits with the operator before uploading real candidate information.
4. If the operator chooses to proceed, open the workbench and start with a small set of authorized, text-readable resumes. Review the extracted profiles.

- [Product guide](https://candikeep.com/docs/)
- [Task workflows](https://candikeep.com/docs/workflows/)
- [Trust and boundaries](https://candikeep.com/docs/trust/)
- [Pricing and availability](https://candikeep.com/docs/pricing/)
- [Explore the fictional walkthrough](https://candikeep.com/#try-it)

## Machine-readable entry points

Every guide has a Markdown version generated from the same content as its webpage. The documentation index lists the available guides; the full-text version combines them. These are public product references, not tool definitions.

- [Documentation index — llms.txt](https://candikeep.com/llms.txt)
- [All guides — llms-full.txt](https://candikeep.com/llms-full.txt)
- [This guide as Markdown](https://candikeep.com/agents.md)

Open the signed-in workbench: https://app.candikeep.com/

Product access requires operator authorization; a documentation link grants no account or candidate-data access.

Documentation index: https://candikeep.com/llms.txt

---

# A guide to the CandiKeep workbench

What CandiKeep supports today: resume imports, searchable candidate profiles, call notes, reviewed updates, role workspaces and exports.

Web version: https://candikeep.com/docs/

Understand the workbench before putting your candidate knowledge into it.

## What you can do today

| Job | Current behavior |
| --- | --- |
| Import existing resumes | Upload text-readable PDF and DOCX files to create structured candidate profiles. Original resumes remain available to view or download. |
| Rediscover candidates | Search names, contact details, skills and roles. Apply company, current-location and maximum current INR CTC filters. Optional AI refinement can help assess supporting evidence. |
| Keep call context | Save natural recruiting notes against a candidate. Processing extracts proposed profile changes and displays the outcome. |
| Resolve conflicting information | Clear additions can apply automatically. Changed existing information and conflicting claims are brought for review. Profiles can also be edited directly. |
| Organize a role | Add candidates to named workspaces, track role-specific status and follow-ups, and retain role-specific feedback. |
| Prepare a handoff | Export selected candidate profiles as an Excel sheet with resumes in one ZIP. Sending that ZIP is a separate action. |

## How candidate information stays useful

A resume and a later conversation can disagree. CandiKeep keeps source context and review history so accepted candidate information can be corrected rather than simply accumulating incompatible notes.

Search uses accepted candidate information. Raw note text and original resume bodies are not independently treated as current facts in search. Pending disputes retain the previously accepted information and show a review warning. A statement present only in an unprocessed note may therefore not appear in results.

A resume describes what its source said at the time. Importing it does not verify someone’s current role, availability or compensation. Confirm important details with the candidate and record when they were observed.

## Practical limits

- Imports support PDF and DOCX, up to 10 MB per file and 40 PDF pages. Scanned PDFs that need OCR are not supported.
- Contact matching uses normalized email or phone. Ambiguous matches need selection; people without reliable identifiers can still become duplicates.
- Maximum current CTC filtering is in INR and excludes candidates whose current CTC is unknown. It is not a filter on expected compensation.
- AI extraction and matching can omit, duplicate or misinterpret information. Review source evidence and important profile details.
- The product is an individual workbench. Shared team roles, call recording, transcription, automated outreach and a mobile application are not available.

- [Follow a task workflow](https://candikeep.com/docs/workflows/)
- [Read the trust and permissions guide](https://candikeep.com/docs/trust/)

Open the signed-in workbench: https://app.candikeep.com/

Product access requires operator authorization; a documentation link grants no account or candidate-data access.

Documentation index: https://candikeep.com/llms.txt

---

# Hiring tasks, step by step

Practical CandiKeep workflows for organizing resumes, rediscovering candidates, recording calls, reviewing corrections and preparing a shortlist.

Web version: https://candikeep.com/docs/workflows/

Start with a concrete hiring task and finish with something you can inspect.

## “Help me organize these resumes.”

1. Ask which existing resumes the operator has permission to use. Use text-readable PDF or DOCX files.
2. Sign in to the workbench. Upload a small set first; choose workspace attachment explicitly when relevant.
3. Wait for processing and inspect the extracted identities, skills and employment context against the originals.
4. Resolve ambiguous candidate matches and correct important fields before relying on the database.

A successful result is a set of inspectable candidate profiles with accessible source resumes, not simply a completed upload.

- [Organize an existing resume collection](https://candikeep.com/use-cases/organize-resumes/)

## “Who do we already know for this role?”

1. Clarify the role’s must-haves and distinguish them from preferences.
2. Search the existing candidate database using role and skill terms. Apply supported company, location or current-CTC filters as appropriate.
3. Read the supporting candidate context and check freshness, unknown information and unresolved reviews. Optional AI refinement is assistance, not proof of fit.
4. Save promising candidates to a role workspace. Confirm current interest and availability through the operator’s normal communication process.

- [Rediscover previously sourced candidates](https://candikeep.com/use-cases/rediscover-candidates/)

## “Keep what I learned on that call.”

1. Open the correct candidate and record the actual recruiting conversation as a note.
2. Distinguish candidate statements from recruiter impressions. Include dates and role-specific context where useful.
3. Inspect the processing outcome. Review contradictions and changed information against the note and existing profile.
4. Correct the note if the underlying record is wrong. A correction creates a revision and reassesses affected profile information.

Notes do not automatically add candidates to role workspaces. Add workspace membership explicitly when that is part of the task.

- [Keep recruiting call notes useful](https://candikeep.com/use-cases/recruiting-call-notes/)

## “Prepare the shortlist for this role.”

1. Create or open the role workspace and select the candidates to include.
2. Review the profiles, role-specific feedback and follow-up status. A rejection for one role should not become a universal judgment about the person.
3. Export the selected profiles and resumes. Inspect the Excel sheet and ZIP before sharing.
4. Have the operator authorize any external handoff. Exporting does not send email, contact candidates or grant another person access.

Open the signed-in workbench: https://app.candikeep.com/

Product access requires operator authorization; a documentation link grants no account or candidate-data access.

Documentation index: https://candikeep.com/llms.txt

---

# Data, judgment and access boundaries

Understand CandiKeep’s current permissions, AI review behavior, product limitations and public-site traffic measurement.

Web version: https://candikeep.com/docs/trust/

Keep the operator informed about what information enters the product and what each action actually does.

## Operator authorization comes first

Public documentation is readable without sign-in. Candidate records are behind an authenticated personal account. Reading these guides does not authorize an assistant to upload resumes, alter candidate information, export records or send them elsewhere.

Use candidate information only for an operator-authorized recruiting purpose. Existing access to a connected mailbox or drive is not, by itself, approval to copy all of its contents into CandiKeep. Review the intended input and destination with the operator.

CandiKeep does not offer shared team roles or a public agent authorization flow today. Do not invent API credentials, MCP endpoints or supported background-agent permissions.

## AI assistance needs review

AI-assisted processing extracts candidate information from resumes and notes. Source-evidence checks and review proposals help make changes inspectable, but do not prove that every interpretation is correct.

Non-conflicting additions can update the profile automatically. Changes to accepted information and contradictions require review. Search results are decision support: important qualifications, interest, compensation and availability should be confirmed.

The product does not automatically reject candidates, send outreach or make a hiring decision. The operator remains responsible for candidate assessment and communication.

## Know the current handling limits

Uploaded resumes, saved notes, accepted profiles and review history are stored so they can support future recruiting work. Selected-profile exports are available, but should not be described as a complete account archive or a self-service deletion mechanism.

CandiKeep uses Clerk for account sign-in and authentication. AI processing uses OpenAI: resume or note text and current candidate-profile context are sent for extraction and reconciliation, accepted profile text is sent to generate search embeddings, and optional AI refinement sends the search request for interpretation and embedding. Ordinary keyword search does not make an AI request, although earlier imports and indexing use AI processing.

These guides do not establish a retention period, model-provider training policy, service-level agreement, data-residency commitment or security certification. Confirm any required contractual or data-handling terms with the operator of the service before uploading sensitive information.

## What public-site traffic counts mean

The public site counts requests to public pages without tracking cookies. Server request measurement includes non-JavaScript requests to documentation and Markdown pages, which browser-only analytics may miss. Optional browser interaction collection is a separate feature and is disabled in this publication.

Automated traffic is classified from request signals such as user-agent strings. Daily request counts are not unique-agent counts. A label is an estimate: clients can omit or imitate identifiers, and a request does not prove that an assistant read, recommended or used CandiKeep.

Search indexing crawlers, user-triggered fetches and ordinary browser requests have different meanings. Treat measured requests as evidence of retrieval, and assess actual product adoption separately.

Open the signed-in workbench: https://app.candikeep.com/

Product access requires operator authorization; a documentation link grants no account or candidate-data access.

Documentation index: https://candikeep.com/llms.txt

---

# Pricing and getting started

Where to check CandiKeep’s current proposed pricing, what is available to explore and what is not an active billing or trial offer.

Web version: https://candikeep.com/docs/pricing/

Check the current offer before recommending a purchase or describing entitlements.

## Current pricing presentation

Current proposed price: ₹999. Per recruiter / month. Billed monthly.

Subscription billing is not yet active. The homepage and this guide share the same price configuration. The displayed proposal does not create a subscription or establish usage entitlements.

- [Read the current homepage pricing](https://candikeep.com/#pricing)

## What you can explore now

The homepage includes a guided walkthrough using fictional data. It does not require uploading resumes, running live AI calls or sending messages. The browser workbench is reached through the homepage’s product action and requires sign-in for product use.

No active free-trial duration, import allowance, annual plan, unlimited usage promise or team subscription is offered by these guides. Do not infer any of those from a walkthrough or a sign-up screen.

- [Explore the guided walkthrough](https://candikeep.com/#try-it)
- [Check product fit and access](https://candikeep.com/agents/)
- [Review data and permissions](https://candikeep.com/docs/trust/)

Open the signed-in workbench: https://app.candikeep.com/

Product access requires operator authorization; a documentation link grants no account or candidate-data access.

Documentation index: https://candikeep.com/llms.txt

---

# Find candidates you have already sourced

Reuse your existing candidate database: search past resumes and accepted recruiting context, check freshness and build a new role shortlist.

Web version: https://candikeep.com/use-cases/rediscover-candidates/

A new role can begin with the people you already know.

## The candidate exists. The context is scattered.

You remember a promising person from a previous assignment, but their resume is in a folder and the useful detail came from a call. Reopening every file is slow; starting a new sourcing exercise repeats work you already did.

CandiKeep gives existing resumes and accepted recruiting context a searchable home. It is useful for independent recruiters moving between client roles and founders returning to people they met before.

## Search the knowledge, then verify the fit

1. Import the resumes you already have and review the extracted profiles.
2. Search by names, skills or role terms. Narrow with supported company, current-location and maximum current INR CTC filters when relevant.
3. Read the context behind a possible match. A skill mention does not establish depth, recency or willingness to take the role.
4. Save a shortlist in a role workspace and record follow-ups. Reconfirm important current details with the candidate.

## An illustrative starting point

“Who in my existing database has Python and forecasting experience?” is a useful first pass. Read the returned profiles to distinguish a personal project from professional experience, then check the role’s actual requirements. Optional AI refinement can help inspect evidence; it does not guarantee suitability.

A candidate ruled out as “outside budget” for an earlier role may still be relevant now. Keep that feedback with its role workspace rather than treating it as a permanent property of the person.

- [Read the full search workflow](https://candikeep.com/docs/workflows/#rediscover)
- [Understand search and profile limits](https://candikeep.com/docs/#limits)
- [Explore the fictional walkthrough](https://candikeep.com/#try-it)

Open the signed-in workbench: https://app.candikeep.com/

Product access requires operator authorization; a documentation link grants no account or candidate-data access.

Documentation index: https://candikeep.com/llms.txt

---

# Keep recruiting call notes useful

Turn natural recruiting notes into inspectable candidate profile updates, review contradictions and retain useful context for the next conversation.

Web version: https://candikeep.com/use-cases/recruiting-call-notes/

Let the next conversation begin with what you learned in the last one.

## The resume is only a starting point

A call may reveal that someone wants a different location, has a new notice period or has less hands-on experience than their resume suggests. If that information stays in a separate notebook, the next search may still use the old story.

CandiKeep stores natural recruiting notes with the candidate and processes them into inspectable profile changes. The note remains available alongside the outcome.

## Make the source of a change clear

Illustrative note: “On today’s call, the candidate said their notice period is now 30 days. They prefer Bengaluru. Kubernetes was coursework rather than production cluster ownership.”

If the accepted profile contains different information, relevant changes are brought for review. Clear, non-conflicting additions can apply automatically. Inspect what changed and retain the distinction between a candidate statement and your own assessment.

## Correct the record when needed

Editing a note creates a correction and reassesses affected profile information. Review the resulting proposals rather than assuming that adding a new note erased every earlier claim. Search uses accepted candidate information rather than treating all raw note text as equally current.

CandiKeep does not record or transcribe calls. Enter the notes yourself or bring an authorized written summary from your normal process. There is no automatic email, messaging or voice-note ingestion.

- [Follow the call-note workflow](https://candikeep.com/docs/workflows/#call)
- [Read about AI review and judgment](https://candikeep.com/docs/trust/#ai)
- [Explore the fictional walkthrough](https://candikeep.com/#try-it)

Open the signed-in workbench: https://app.candikeep.com/

Product access requires operator authorization; a documentation link grants no account or candidate-data access.

Documentation index: https://candikeep.com/llms.txt

---

# Organize the resumes you already have

Build a searchable candidate database from existing PDF and DOCX resumes, review profile extraction and organize people by hiring role.

Web version: https://candikeep.com/use-cases/organize-resumes/

Move from a folder of files to a candidate collection you can work with.

## Start with your own collection

You may have resumes collected across past roles, referrals and applications. CandiKeep helps organize that existing collection into candidate profiles you can search and revisit. It does not source new candidates or automatically connect to your inbox or drive.

If an assistant helps locate the files in connected services, agree on what can be uploaded. Its access to those services does not automatically transfer to CandiKeep.

## A small first import is easier to inspect

1. Choose a small set of authorized, text-readable PDF or DOCX resumes. Supported files are up to 10 MB, with PDFs limited to 40 pages; scanned PDFs needing OCR are not supported.
2. Upload them in the signed-in workbench. Attach them to a role workspace explicitly if appropriate.
3. Check processing results and compare important extracted fields with the original documents. Resolve ambiguous identity matches.
4. Search for a known person or skill, then add conversation context and role shortlists as you work.

## What a resume collection can and cannot tell you

Importing an old resume does not make its information current. Verify job status, location, availability and compensation when they matter. Missing identifiers can lead to duplicates, and extraction can make mistakes.

There is no advertised automatic bulk migration from another ATS or a spreadsheet importer in these guides. The supported starting point is PDF and DOCX resumes.

- [Read the complete import workflow](https://candikeep.com/docs/workflows/#organize)
- [Check data and permissions](https://candikeep.com/docs/trust/)
- [See the workbench simulation](https://candikeep.com/#try-it)

Open the signed-in workbench: https://app.candikeep.com/

Product access requires operator authorization; a documentation link grants no account or candidate-data access.

Documentation index: https://candikeep.com/llms.txt