Dealmachine
Search and enrich US property, owner, and people data with DealMachine through its MCP tools or dm CLI. Use for real estate lead generation, property and owner lookup, contact enrichment, prospecting, comparable sales, market research, targeted lists or exports, and credit-aware DealMachine workflows.From its SKILL.md
npx -y skills add DealMachine/dealmachine-cli --skill dealmachineAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its file declares
Copied from the file, not written here
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
6.7 KB, as published. Nobody here has run it
DealMachine
Use DealMachine to turn a user's sales, marketing, lead generation, or real estate research request into a precise property, people, enrichment, comparable-sales, or export workflow.
Choose the interface
- Prefer the
dealmachine_*MCP tools when they are available. - Otherwise, use the
dmCLI if shell access is available. - If neither interface is connected, help the user connect the hosted MCP server at
https://mcp.dealmachine.comor install the CLI withnpm install -g dealmachine, then rundm login.
Never ask the user to paste an API key into chat.
Read REFERENCE.md when you need the exact MCP Tool ID, CLI command group, or current interface coverage. The reference tracks DealMachine CLI and hosted MCP 0.3.0.
Protect credits
- Treat filters, fields, locations, usage, identity checks, and count operations as discovery steps.
- Discover valid filter IDs and field IDs before searching. Never invent them.
- Count or estimate the matching audience before a paid search or export.
- Explain the expected result count, fields, contact audience, and credit impact before a large paid operation.
- Ask for confirmation before an export or an operation that can consume a material number of credits.
- For property-only requests, do not add owner or resident contact enrichment.
- For explicit single-record lookups, run the requested lookup and report whether credits were used.
- For CLI property search, people search, and name enrichment, run the free
--estimate-coststep first. Add--yesonly after the user approves the displayed estimate.
Require approval for external effects
- Immediately before any operation that can consume credits, create an export, write or delete data, send mail, change account settings, or initiate a purchase, state the exact effect and obtain explicit user approval.
- A broad request to "handle it" is not approval for an irreversible action. Prefer a count, cost estimate, preview, or dry run first.
- Never use the plugin or CLI to open checkout, add funds, change a subscription, or purchase digital credits. Explain the limitation and point to DealMachine account settings only when the user asks.
- Do not execute task, list, direct-mail, account, developer-license, or other CLI write commands from this skill. Limit the CLI fallback to the research workflows documented below.
Plan the request
Identify:
- Entity: property, person, or list
- Location: address, city, county, ZIP code, state, coordinates, or APN
- Criteria: ownership, equity, property characteristics, demographics, or other filters
- Output: count, preview, detailed results, enrichment, comparable sales, list, or export
- Fields: only the data needed for the task
- Contact audience: owners, residents, both, or none
- Limit and format: a small preview first unless the user specifies otherwise
Ask one focused question only when a missing detail would materially change the search or credit cost.
Use MCP tools
Discovery
- Use
dealmachine_filtersto find valid property or people filters. - Use
dealmachine_fieldsto find valid output fields. - Use
dealmachine_location_searchto resolve supported locations. - Use
dealmachine_usagebefore large paid operations. - Use
dealmachine_whoamiwhen account context matters.
Property workflows
- Use
dealmachine_property_countbefore a broad search. - Use
dealmachine_property_searchfor a limited result set. - Use
dealmachine_property_getordealmachine_property_get_manyfor known IDs. - Use
dealmachine_property_exportonly after confirming scope and credit impact. - Use
dealmachine_compsfor comparable sales.
People workflows
- Use
dealmachine_people_countbefore a broad search. - Use
dealmachine_people_searchfor results and the matching get tools for known IDs. - People search and lookup are available through MCP. Use the CLI for a people export.
- A specific person by name is enrichment, not People Search. Use
dealmachine_enrich_nameordm enrich name, narrowed by state, ZIP code, county, or city place ID when available.
Enrichment
- Use the address, latitude and longitude, or APN enrichment tool for a property.
- Use the email, phone, or name enrichment tool for a person.
- Narrow name searches with a location.
Use the CLI
Run dm --help or a command-specific --help when exact syntax is uncertain.
Common patterns:
dm filters --source-type properties --search "absentee owner" --json
dm fields --source-type properties --search "equity" --json
dm properties count --body-file search.json --json
dm properties search --body-file search.json --estimate-cost --json
dm properties search --body-file search.json --json --yes
dm enrich address "123 Main St, Austin, TX" --json
dm enrich phone "5125550100" --json
dm enrich name "Jane Smith" --state TX --estimate-cost --json
dm enrich name "Jane Smith" --state TX --json --yes
dm comps PROPERTY_ID --json
dm usage --json
Use JSON output for automation. Store complex request bodies in a temporary JSON file instead of constructing fragile shell-escaped payloads.
Use the CLI fallback only for the property, people, enrichment, comparable-sales, discovery, count, and explicitly approved export workflows documented in REFERENCE.md.
Present the result
- State what was searched and which filters were applied.
- Distinguish estimates, counts, previews, and complete exports.
- Report records returned and credits used when the response provides those values.
- Show the most decision-useful fields first.
- Preserve uncertainty when a field is missing or inferred.
- Offer the next logical action, such as narrowing the audience, enriching selected records, finding comps, or exporting the confirmed set.
Boundaries
- Do not claim a phone number, email, valuation, ownership fact, or other attribute that the result does not contain.
- Do not describe a count as a completed export.
- Do not expose access tokens, API keys, or raw credentials.
- Do not bypass DealMachine authentication, plan limits, or credit controls.