merge-contacts
Skill34 starsChanged 24 days agoMITView on GitHub
Merge the user's contact exports (Google Takeout, LinkedIn export, vault People notes) into one local CRM. Use for "/merge-contacts", "build my CRM", "dedup my contacts", or when another workflow needs a clean contacts source of truth.
In this repository21 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- add-agentSkill
- build-listSkill
- chart-tufteSkill
- councilSkill
- design-precheckSkill
- document-exportSkill
- handoverSkill
- index-memorySkill
- log-decisionSkill
- mcp-debugSkill
- meeting-prepSkill
- memory-debtSkill
- morning-briefSkill
- post-callSkill
- reflectSkill
- setupSkill
- system-auditSkill
- tasksSkill
- visual-explainerSkill
- weekly-reviewSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Find the exports: Check, in order:
- Very specific: line 20
If none exist, point the user at the setup steps in modules/crm-merge/README.md (both exports are self-service downloads) and stop.
- Very specific: line 45
Over-merging on a common name: if the user says distinct people got fused, lower --name-collision-threshold and re-run.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Merge Contacts
Build the user's local CRM source of truth with the crm-merge module. The
module does the work; your job is to find their exports, run it with the right
flags, and report what merged. Never invent data, never call an external CRM.
Steps
- Find the exports. Check, in order:
- A Google Takeout zip in
~/Downloads(name contains "takeout" or "google"). The script auto-finds the newest one, but confirm it exists. - A LinkedIn export folder (contains
Connections.csv, ideallymessages.csv). Ask the user for the path if you cannot find it. vault/People/for per-person notes (optional). If none exist, point the user at the setup steps inmodules/crm-merge/README.md(both exports are self-service downloads) and stop.
- A Google Takeout zip in
- Get their name for message-frequency counting. Ask how their name appears
in LinkedIn messages (the "me" side), or read it from
config/user.config.yaml. - Run the build:
Omit any source they do not have; one source alone is fine.python3 modules/crm-merge/build.py \ --takeout-zip <path-or-omit-to-autofind> \ --linkedin-dir <folder> \ --me "<their name>" \ --out data/crm-merge - Report the counts the script prints: total unique people, how many appear in 2+ sources (the high-confidence merges), and how many link to a vault note. Flag if a source loaded zero records, since that usually means a wrong path.
- Point them onward. Tell them
data/crm-merge/browser.htmlis openable in any browser, and thatcontacts.jsonis what/build-listdedups against.
Watch out for
- Over-merging on a common name: if the user says distinct people got fused,
lower
--name-collision-thresholdand re-run. - Zero LinkedIn messages counted: the
--mename probably does not match the export. Check theFROM/TOvalues inmessages.csvand pass the exact form. - Privacy: outputs are git-ignored; do not commit them or paste contact data into anything outward-facing.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

