# AI use cases for CRM (trycompai/crm) Repo: [trycompai/crm](https://github.com/trycompai/crm) | Use cases: 4 | Page: https://dedupedai.com/use-cases/trycompai-crm/ Comp AI CRM is an open source customer relationship manager built so that AI agents do the research and record what they find. This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL. - Run the CRM locally with a sample pipeline: https://dedupedai.com/use-cases/trycompai-crm/run-comp-crm-locally-with-seed-data/ - Have an agent fill contact gaps from your own email threads: https://dedupedai.com/use-cases/trycompai-crm/let-agent-enrich-contacts-from-crm-history/ - Pull brand data and a LinkedIn profile into a prospect record: https://dedupedai.com/use-cases/trycompai-crm/research-a-prospect-company-with-context-data/ - Audit what the agent did on a deal before you trust it: https://dedupedai.com/use-cases/trycompai-crm/review-agent-steps-and-decisions-on-a-deal/ ## How to run the CRM locally with a sample pipeline with CRM Repo: [trycompai/crm](https://github.com/trycompai/crm) | Difficulty: Intermediate | Time: 1 hour | Basis: From the README Clone the repo, start Postgres in Docker, and open a seeded CRM on your own machine. **Who it is for:** A developer or sales ops lead who wants to see how the CRM and its agent behave before deploying anything. **The problem:** Reading about an agent-first CRM does not show how the records, tabs and pipeline look in use. **The outcome:** The app runs on localhost:3000 with a believable pipeline loaded, and the API runs on localhost:3001. ### Steps 1. Run `git clone https://github.com/trycompai/crm.git && cd crm` 2. Run `cp .env.example .env`, then set BETTER_AUTH_SECRET, ALLOWED_SIGN_IN and at least one of Google or Microsoft sign-in values 3. Run `bun install` 4. Run `docker compose up -d`, then `bun run db:deploy` and `bun run db:seed` 5. Run `bun run dev` and open http://localhost:3000 ### Prompt to paste into your agent ```text Clone the trycompai/crm repo, copy .env.example to .env, set BETTER_AUTH_SECRET and ALLOWED_SIGN_IN, then run bun install, docker compose up -d, bun run db:deploy, bun run db:seed and bun run dev. Report the URL that serves the app. Verify that http://localhost:3001 responds before you finish. ``` ### Caveat Sign-in needs a Google or Microsoft OAuth client, and the README warns that the Google client should be set up in pairs or not at all. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/trycompai-crm/run-comp-crm-locally-with-seed-data/ ## How to have an agent fill contact gaps from your own email threads with CRM Repo: [trycompai/crm](https://github.com/trycompai/crm) | Difficulty: Advanced | Time: 2 hours | Basis: From the README Run the research agent so it reads your CRM's threads, meetings and signature blocks and records strong facts. **Who it is for:** A sales rep or RevOps lead with many contacts that have missing job titles or companies after email and meetings have already happened. **The problem:** Contact records go stale or stay blank because nobody updates them after each conversation. **The outcome:** The agent records facts it can support from CRM history, and weak evidence appears as suggestions for a person to settle. ### Steps 1. Deploy apps/agent next to the CRM and set AGENT_BRIDGE_SECRET to the same value in both processes 2. Open a contact and select the Agent tab to see the steps the agent takes 3. Review the startup list that shows which sources are on (for example, Web research with PERPLEXITY_API_KEY) 4. Check the facts it recorded on the contact and resolve any suggestions it could not decide ### Prompt to paste into your agent ```text In the trycompai/crm repo, use the apps/agent research agent to run read_crm_history and record_fact for the contact , with no outside API keys set. Report the facts recorded and any suggestions left for a human. Verify that every recorded fact cites an observed source such as a signature block or thread. ``` ### Caveat The README says the agent is designed to run with no outside keys, but it still needs a deployed apps/agent and a running CRM, and the README does not give the exact command to start the agent. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/trycompai-crm/let-agent-enrich-contacts-from-crm-history/ ## How to pull brand data and a LinkedIn profile into a prospect record with CRM Repo: [trycompai/crm](https://github.com/trycompai/crm) | Difficulty: Intermediate | Time: 45 minutes | Basis: From the README Add a Context key in Settings → General so accounts show real logos, names and LinkedIn roles. **Who it is for:** An account executive preparing for a first call who wants a company and person to look right in the CRM before they meet. **The problem:** Accounts arrive as grey squares with initials, and a person's address does not reveal who they are. **The outcome:** The company shows its real name, logo, colours and industry, and a contact with a LinkedIn URL on record gets a verified name, title and employer history. ### Steps 1. Sign in to the CRM and open Settings → General 2. Enter your Context key in the LinkedIn and company brand data field 3. Confirm the startup list shows Company brand data and LinkedIn as on 4. Open a prospect company and a contact that has a LinkedIn URL and check the updated fields ### Prompt to paste into your agent ```text In the trycompai/crm CRM, confirm that the Context key is set under Settings → General, then trigger enrichment for the company and the contact . Report the logo, industry, current title and employer that were recorded. Verify that the startup list shows both Company brand data and LinkedIn as on. ``` ### Caveat The README says the LinkedIn lookup works from a LinkedIn URL already on the record, so contacts without one get no LinkedIn data. Risk, credentials: The Context key controls paid lookups against people's LinkedIn data, so store it only in Settings → General and never in code or chat. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/trycompai-crm/research-a-prospect-company-with-context-data/ ## How to audit what the agent did on a deal before you trust it with CRM Repo: [trycompai/crm](https://github.com/trycompai/crm) | Difficulty: Beginner | Time: 30 minutes | Basis: From the README Open the Agent tab on a deal to see each step, discarded leads and open questions. **Who it is for:** A sales manager reviewing a pipeline that an agent has been updating on its own schedule. **The problem:** Automated changes to records are hard to trust when nobody can see how a conclusion was reached. **The outcome:** The manager can see each step, why leads were dropped, and answer any question the agent could not settle. ### Steps 1. Open a deal in the CRM and select the Agent tab 2. Read the steps the agent took and the leads it threw away with reasons 3. Answer any open question the agent asked about two people who might match 4. Check that schedule_recheck entries show the reason the agent gave for returning ### Prompt to paste into your agent ```text In the trycompai/crm CRM, open the Agent tab for the deal , list each step, each discarded lead with its reason, and each open question. Report those items in a short table. Verify that every schedule_recheck entry includes a stated reason. ``` ### Caveat The Agent tab only shows conversations after AGENT_BRIDGE_SECRET is set in both processes, and otherwise reports that it is not configured. Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/ URL: https://dedupedai.com/use-cases/trycompai-crm/review-agent-steps-and-decisions-on-a-deal/