메뉴
Binder CLI for knowledge graph operations — CRUD, search, schema inspection, transaction import, docs rendering. Use when asked to "query binder", "search records", "create a record", "check the schema", "import transactions", "undo changes", or work with a binder workspace.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Binder data modeling — define entity types, fields, relations, constraints, views, and navigation. Use when asked to "create a type", "add a field", "define a schema", "set up relations", "model entities", "create a view", "set up navigation", "render entities as files", or design a binder workspace schema.
Quick timestamped log entry — no conversation, just capture.
Build a day plan with carry-forwards and one critical item.
Day briefing — deliver context, capture what's happened, get moving.
Evening summary — reflect on the day, score it, close it out.
Period planning (w/m/q/y) — set goal and plan for the next period.
| name | binder-cli |
| description | Binder CLI for knowledge graph operations — CRUD, search, schema inspection, transaction import, docs rendering. Use when asked to "query binder", "search records", "create a record", "check the schema", "import transactions", "undo changes", or work with a binder workspace. |
Binder is a Markdown-native knowledge graph. The CLI is the primary interface for all operations.
Run binder --help or binder <command> --help for full usage details.
binder read <ref> # Read a record
binder search type=Task status=active # Search records
binder schema --types Task # Inspect schema
binder create <Type> field=value [field=value...] # Create a record
binder update <ref> field=value [field=value...] # Update fields
binder delete <ref> # Delete a record
binder tx import -d file.yaml # Dry-run a transaction file
binder tx import file.yaml # Apply a transaction file
binder tx log --limit 5 # Recent transactions
binder undo [N] # Undo last N transactions
Global options: -C <path> cwd, -q quiet, --format output format, -n namespace.
binder read <ref> [--format yaml|json] [-f fields]
Ref: id (numeric), uid (e.g. tsk-abc123), or key (e.g. Task, status).
Use -f/--fields to include or traverse relations inline — same syntax as search.
binder search type=Task status=active # Filter by field values
binder search "some text" # Full-text search
binder search type=Task -f partOf(title) # Include fields from related records
binder search type=Task -o '!priority' # Order by (! = descending)
binder search type=Task --limit 10 # Limit results
Options:
--format — json, jsonl, yaml, csv, tsv-n, --namespace — record (default) or config-f, --fields — include specific fields or traverse relations. title,status includes those fields. partOf(title,status) follows the partOf relation and includes title and status from the related record. Use relatesTo[type=Task](title) to filter relation targets by type before traversal.-o, --orderBy — sort (prefix ! for descending)--limit — max resultsbinder schema # All types and fields
binder schema --types Task # Specific type(s)
binder schema -n config # Config namespace
binder create <Type> field=value [field=value...] # Create a record
binder update <ref> field=value [field=value...] # Update fields
binder delete <ref> # Delete a record
For 3+ changes or mixed namespace operations, use transaction files instead.
field=value — setfield+=value — append to arrayfield-=value — remove from arrayfield:0+=value — insert at positionfield:last-- — remove last'fields:title={required: true}' — patch nested attributesDelimiter depends on the field's text format (check binder schema):
field=a,b,cfield=$'a\nb\nc'field=$'first block\n\nsecond block'Set a relation field to the target's uid or key:
binder create Task key=fix-auth title="Fix auth" taskType=fix partOf=mst-v1
binder update tsk-abc123 requires+=tsk-def456 # Add a relation
binder update tsk-abc123 requires-=tsk-def456 # Remove a relation
binder update tsk-abc123 relatesTo=tsk-x,tsk-y # Set multiple relations
The target record must exist before you can reference it. Use binder search to find the target's uid first.
For bulk or multi-step changes, write a transaction YAML file.
Workflow: write YAML → dry-run (binder tx import -d file.yaml) → show output to user and ask for approval → apply (binder tx import file.yaml). Always dry-run first, never skip approval.
Each entry has author and records/configs arrays. Use type to create, $ref (uid or key) to update:
- author: agent
records:
- type: Task
title: Implement auth
status: pending
partOf: mst-v1
requires:
- tsk-def456
- $ref: tsk-abc123
status: done
tags:
- - insert
- urgent
configs:
- $ref: nav-backlog
where:
type: Task
status: pending
List field mutations in transaction files: ["insert", value], ["insert", value, position], ["remove", value], ["patch", ref, {attrs}].
binder tx log # Recent transactions
binder tx log --limit 5 # Last N
binder tx read <ref> # Read a transaction
binder undo [N] # Undo last N transactions
binder redo [N] # Redo last N undone
binder search type=Task key=<key>binder docs render # Render navigation to files
binder docs sync # Sync docs with knowledge graph
binder docs lint # Validate YAML and Markdown files
binder locate <ref> # Print file path and line number
binder init # Initialize new workspace
binder schema if it is not in your context--format tsv for scannable search results--format yaml for single records and nested data--format json | jq for complex data processing