Skip to main content Skills Marketplace Discover and explore AI skills built by the community.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Show prompt details A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/dzhng/crm.cli --skill crm-cliThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Download Zip Downloading...
Related occupations SOC
Based on SOC occupation classification
name crm-cli description Manage contacts, companies, deals, and pipeline with crm.cli — a headless CLI-first CRM backed by SQLite with a virtual filesystem interface install curl -fsSL https://raw.githubusercontent.com/dzhng/crm.cli/main/install.sh | sh
crm.cli
A headless, CLI-first CRM. Contacts, deals, and pipeline in a single SQLite file — queryable from your terminal, composable with Unix tools, and mountable as a virtual filesystem.
Install
curl -fsSL https://raw.githubusercontent.com/dzhng/crm.cli/main/install.sh | sh
This downloads the precompiled binary to ~/.local/bin and installs mount dependencies (FUSE on Linux, Rust toolchain on macOS for NFS).
After install, make sure ~/.local/bin is in your PATH:
export PATH="$HOME /.local/bin:$PATH "
Verify:
crm --version
Configuration
Optional. Create crm.toml in your project root or ~/.crm/config.toml:
[database]
path = "~/.crm/crm.db"
[pipeline]
stages = ["lead" , "qualified" , "proposal" , "negotiation"
"closed-won"
"closed-lost"
won_stage
"closed-won"
lost_stage
"closed-lost"
[defaults]
format
"table"
[phone]
default_country
"US"
display
"international"
[mount]
default_path
"~/crm"
Config is auto-discovered by walking up from the current directory. Override with --config <path> or CRM_CONFIG env var.
Global Flags
--db <path> — SQLite database path (default: ~/.crm/crm.db, env: CRM_DB)--format <fmt> — Output format: table, json, csv, tsv, ids--config <path> — TOML config file path--no-color — Disable colored output
Contacts
Create a contact crm contact add --name "Jane Doe" \
--email jane@acme.com \
--phone "+1-212-555-1234" \
--linkedin linkedin.com/in/janedoe \
--company "Acme Corp" \
--tag hot-lead \
--set title=CTO
All flags are optional except --name. Phones are normalized to E.164, LinkedIn URLs are extracted to handles, companies are auto-created if they don't exist. --email, --phone, --company, --tag, and --set are all repeatable.
Social handle flags: --linkedin, --x, --bluesky, --telegram. All accept raw handles or full URLs.
List contacts crm contact list
crm contact list --tag hot-lead --company "Acme Corp"
crm contact list --filter "title~=CTO AND company=Acme" --sort name --limit 20
crm contact list --format json | jq '.[].name'
Filter operators: =, !=, ~= (contains), >, <. Combine with AND / OR.
Show a contact Look up by ID, email, phone, or social handle:
crm contact show ct_01J8ZVXB3K...
crm contact show jane@acme.com
crm contact show "+12125551234"
crm contact show janedoe
Edit a contact crm contact edit jane@acme.com --name "Jane Smith"
crm contact edit "+12125551234" --add-email jane2@acme.com --rm-tag old-tag
crm contact edit janedoe --add-company "New Corp" --set title=CEO --unset source
Add/remove flags: --add-email, --rm-email, --add-phone, --rm-phone, --add-company, --rm-company, --add-tag, --rm-tag. Social handles set directly: --linkedin, --x, --bluesky, --telegram.
Delete a contact crm contact rm jane@acme.com
crm contact rm "+12125551234" --force
crm contact rm janedoe
Merge contacts Merge two contacts into one. First contact survives, second is absorbed and deleted. Accepts any reference type:
crm contact merge ct_01A... ct_01B...
crm contact merge jane@acme.com jane.doe@acme.com
crm contact merge "+12125551234" "+14155559876"
crm contact merge janedoe jane-doe-linkedin
Combines emails, phones, companies, tags, custom fields, and relinks all deals and activity.
Companies
Create a company crm company add --name "Acme Corp" \
--website acme.com \
--phone "+1-800-555-0000" \
--tag enterprise \
--set industry=SaaS
--website, --phone, --tag, --set are repeatable.
List / show / edit / delete crm company list --tag enterprise
crm company show acme.com
crm company show "+18005550000"
crm company edit acme.com --name "Acme Inc" --add-website acme.io
crm company rm acme.com --force
Merge companies crm company merge co_01A... co_01B...
crm company merge acme.com acme.io
crm company merge "+18005550000" "+18005550001"
Relinks all contacts and deals from second to first.
Deals
Create a deal crm deal add --title "Acme Enterprise" \
--value 50000 \
--stage qualified \
--contact jane@acme.com \
--company acme.com \
--expected-close 2026-06-15 \
--probability 60 \
--tag enterprise
Contacts and companies are auto-created if they don't exist. --contact and --tag are repeatable.
List deals crm deal list --stage qualified --min-value 10000
crm deal list --contact jane@acme.com --sort value --reverse
crm deal list --format ids | wc -l
Move a deal through the pipeline crm deal move dl_01... --stage proposal --note "Sent pricing deck"
crm deal move dl_01... --stage closed-won --note "Signed 2-year contract"
Stage transitions are recorded as activity with timestamps. Use deal move, not deal edit --stage.
Edit / delete crm deal edit dl_01... --value 75000 --add-contact bob@acme.com --probability 80
crm deal rm dl_01... --force
Pipeline overview Shows count, total value, and weighted value per stage.
Activity Logging
Log an activity crm log note "Had coffee with Jane, discussed Q3 expansion" --contact jane@acme.com
crm log call "Demoed product, she wants a proposal" --contact jane@acme.com --deal dl_01...
crm log meeting "Quarterly review" --company acme.com --at 2026-04-01
crm log email "Sent follow-up pricing" --contact jane@acme.com --set channel=outbound
Types: note, call, meeting, email. Contacts and companies are auto-created. --contact is repeatable. --at overrides the timestamp.
List activities crm activity list --contact jane@acme.com --since 2026-01-01
crm activity list --type call --limit 10
crm activity list --deal dl_01... --format json
Tags crm tag jane@acme.com hot-lead enterprise
crm untag jane@acme.com old-tag
crm tag list
crm tag list --type contact
Tags work on contacts, companies, and deals.
Search
Exact keyword search (FTS5) crm search "acme CTO"
crm search "jane" --type contact
Fuzzy / semantic search crm find "fintech startup London"
crm find "that CTO I met at the conference" --limit 5 --threshold 0.3
Rebuild search index crm index rebuild
crm index status
Index updates automatically on writes. Manual rebuild only needed after corruption.
Duplicate Detection crm dupes
crm dupes --type contact --threshold 0.5
crm dupes --type company --limit 20
Uses combined Levenshtein + Dice coefficient similarity. Detects: similar names, shared emails, shared phones, shared websites, shared social handles. Review then merge:
crm dupes --type contact
crm contact merge ct_01A... ct_01B...
Reports crm report pipeline
crm report activity --period 30d --by type
crm report stale --days 14 --type contact
crm report conversion --since 2026-01-01
crm report velocity --won-only
crm report forecast --period 2026-Q2
crm report won --period 90d
crm report lost --period 90d
Import / Export
Import from CSV or JSON crm import contacts leads.csv
crm import contacts leads.json --update
crm import companies companies.csv --dry-run
crm import deals deals.csv --skip-errors
cat data.json | crm import contacts -
CSV headers: name, email/emails, phone/phones, company/companies, tags, linkedin, x, bluesky, telegram. Unrecognized columns become custom fields.
Export crm export contacts --format csv > contacts.csv
crm export companies --format json > companies.json
crm export deals --format tsv
crm export all --format json > full-backup.json
Virtual Filesystem (Mount) Mount the CRM as a live read/write filesystem. Any tool that reads files gets full CRM access — AI agents, grep, jq, vim, scripts.
Mount crm mount ~/crm
crm mount ~/crm --readonly
On Linux this uses FUSE. On macOS this uses an NFS v3 server (no kernel extensions needed).
Filesystem layout ~/crm/
├── llm.txt # Instructions for AI agents
├── contacts/
│ ├── ct_01...jane-doe.json # Contact JSON files
│ ├── _by-email/ # Lookup by email
│ ├── _by-phone/ # Lookup by E.164 phone
│ ├── _by-linkedin/ # Lookup by LinkedIn handle
│ ├── _by-x/ # Lookup by X handle
│ ├── _by-company/ # Grouped by company
│ └── _by-tag/ # Grouped by tag
├── companies/
│ ├── co_01...acme-corp.json
│ ├── _by-website/
│ ├── _by-phone/
│ └── _by-tag/
├── deals/
│ ├── dl_01...acme-enterprise.json
│ ├── _by-stage/
│ ├── _by-company/
│ └── _by-tag/
├── activities/
│ ├── _by-contact/
│ ├── _by-company/
│ ├── _by-deal/
│ └── _by-type/
├── reports/ # Pre-computed analytics
│ ├── pipeline.json
│ ├── forecast.json
│ ├── stale.json
│ ├── conversion.json
│ ├── velocity.json
│ ├── won.json
│ └── lost.json
├── pipeline.json # Quick pipeline overview
├── tags.json # All tags with counts
└── search/ # Search by reading files
└── <query>.json # cat search/"acme CTO".json
Read via filesystem ls ~/crm/contacts/
cat ~/crm/contacts/ct_01...jane-doe.json | jq .
cat ~/crm/contacts/_by-email/jane@acme.com.json
cat ~/crm/deals/_by-stage/qualified/
cat ~/crm/reports/forecast.json
cat ~/crm/search/"enterprise deals" .json
Write via filesystem
echo '{"name":"Bob Smith","emails":["bob@globex.com"]}' > ~/crm/contacts/new.json
cat ~/crm/contacts/ct_01...jane-doe.json | jq '.tags += ["vip"]' > ~/crm/contacts/ct_01...jane-doe.json
rm ~/crm/contacts/ct_01...jane-doe.json
Unmount
Static export (no mount needed) crm export-fs ./crm-snapshot
Exports the same directory structure as a static copy — useful in containers or sandboxes where FUSE isn't available.
Bulk Operations Use --format ids to pipe into other commands:
crm contact list --company "Acme Corp" --format ids | xargs -I{} crm tag {} enterprise
crm deal list --stage qualified --min-value 50000 --format ids | \
xargs -I{} crm deal move {} --stage proposal
crm report stale --days 90 --type contact --format ids | xargs -I{} crm contact rm {} --force
Custom Fields All entities support arbitrary key-value fields:
crm contact add --name "Jane" --set title=CTO --set source =conference
crm contact edit jane@acme.com --set "json:score=85" --set "json:verified=true"
crm contact edit jane@acme.com --unset source
crm contact list --filter "title~=CTO"
Prefix with json: for typed values (numbers, booleans, arrays).
Hooks Configure shell hooks in crm.toml that fire on mutations:
[hooks]
post-contact-add = "~/.crm/hooks/notify-slack.sh"
post-deal-stage-change = "~/.crm/hooks/deal-moved.sh"
pre-contact-rm = "~/.crm/hooks/confirm-delete.sh"
Entity data is passed as JSON on stdin. Pre-hooks abort on non-zero exit.
Available hooks: {pre,post}-{contact,company,deal}-{add,edit,rm}, {pre,post}-deal-stage-change, {pre,post}-activity-add.
Tips for AI Agents
Mount first: crm mount ~/crm gives you filesystem access — read JSON files directly instead of running CLI commandsRead llm.txt: The mount point contains llm.txt with structure docs and tipsUse _by-* directories for fast lookups: _by-email, _by-phone, _by-linkedin, _by-tag, _by-stageUse --format json for all CLI output when processing programmaticallyUse --format ids + xargs for bulk operationsRead reports/ for pre-computed analytics — don't recompute from raw dataSearch via filesystem: cat ~/crm/search/"your query".jsonWrite via filesystem: Create/update entities by writing JSON filesAll JSON files are self-contained — no need to join across filesPhone numbers accept any format on input; stored as E.164 internallySocial handles accept full URLs; stored as clean handles 