OnLink Bulk Job: Teamwork Graph CSV Export from Any HR Source

Use this guide to configure an OnLink Teamwork Graph bulk job. The job pulls employee/org data from any HR connection (Workday, BambooHR, ADP, Anysource, etc.), maps it to Atlassian’s HRIS S3 format, and uploads a CSV to your S3 bucket for the Teamwork Graph HRIS S3 connector.

What it does

Each run:

Fetches all records from your Source HR connection
Maps source fields → Teamwork Graph fields (Workers, Positions, or Organizations)
Normalizes values (dates, booleans, enums) to Atlassian’s expected format
Builds a full-snapshot CSV and overwrites the fixed file name in your S3 bucket

Object type	CSV file uploaded

Object type	CSV file uploaded
Workers	`workers.csv`
Positions	`positions.csv`
Organizations	`organizations.csv`

You typically need one bulk job per object type (three jobs for a full org chart).

Atlassian field reference:
https://support.atlassian.com/organization-administration/docs/hris-s3-connector-field-reference/

Prerequisites

In Atlassian

Teamwork Graph HRIS S3 connector configured to read from your S3 bucket
Connector expecting the standard file names above

In OnLink

HR Source connection (any saved connection)
Two identities (simplest and matches how each product works today):
- IAM user + access key → OnLink write (PutObject)
- IAM role + External ID → Atlassian read (GetObject / ListBucket)

Examples: Workday, BambooHR, Oracle HCM, Azure, Anysource (custom REST API).

S3 Target connection

OnLink → Connections → Add System → HRIS S3

Create the bulk job

OnLink → Bulk Jobs → Add New Bulk Job

Field	What to set

Field	What to set
Batch job name	Descriptive name, e.g. `TG Workers – Workday`
Bulk job type	Teamwork Graph
Object type	Workers, Positions, or Organizations
Source	HR connection that supplies the data
Target	AWS S3 Anysource connection (SigV4)
Schedule	Optional — toggle on for daily/hourly/weekly/monthly runs

Teamwork Graph jobs use a single screen (no JQL / HTTP body step). Click Submit when done.

Configure field mapping

The mapping editor pre-fills a template with one line per Teamwork Graph field:

map:<source field name>=workerId
map:<source field name>=createdAt
...

Replace <source field name> with the flattened field name from your HR API (use Fetch API Sample — see below).

Mapping line format

Type	Format	Example

Type	Format	Example
Field map	`map:<sourceField>=<targetField>`	`map:employee_id=workerId`
Fetch config	`config:<key>=<value>`	`config:data_path=results`

Supported `config:` keys (Anysource / API tuning)

Key	Purpose

Key	Purpose
`api`	Override API path
`data_path`	JSON path to the record array
`http_body`	POST request body (JSON)
`http_method`	HTTP method override
`max_pages`	Cap pagination pages
`connection_name`	Anysource connection name (usually auto-filled)

Note: Only one : per line is allowed in the mapping parser. Custom HTTP headers are not supported in bulk-job mappings.

Required fields (must be mapped)

Workers: workerId, createdAt, name, isActive

Positions: positionId, createdAt, isActive, organizationId, workerId

Organizations: organizationId, organizationName, createdAt, isActive

Optional fields can be left as placeholders; they export as empty columns.

Test before saving

Select Source and enter mappings
Click Fetch API Sample
Review Sample Data — confirms connectivity and shows exact source field names for map: lines
Fix mapping until required fields are covered
Click Submit

Run the job

Manual run

Bulk Jobs list → Run on the job row.

Scheduled run

Enable Schedule, set period (daily / hourly / weekly / monthly), save.

Value formats (automatic normalization)

OnLink converts mapped values before writing CSV:

Field kind	Output format	Examples accepted

Field kind	Output format	Examples accepted
datetime	ISO UTC, e.g. `2024-09-18T13:00:00.000Z`	Most date strings
date	`YYYY-MM-DD`	Most date strings
bool01 (`isActive`)	`1` or `0`	true/false, yes/no, active/inactive
enum	Canonical value	`status`: Filled / Unfilled; `employmentType`: Employee / Contingent / Contractor

Rows missing required fields are dropped. Invalid optional values are blanked.

You can use below transformations:

concat(firstName, ‘ ‘, lastName) e.g. map:concat(firstName, ‘ ‘, lastName)=name
if(status eq ’active’,’1’,’0’) e.g. map:if(status eq ’active’,’1’,’0’)=isActive

Recommended setup (full org chart)

Create three bulk jobs sharing the same Source and Target:

Job name	Object type	Output file

Job name	Object type	Output file
TG Organizations	Organizations	`organizations.csv`
TG Positions	Positions	`positions.csv`
TG Workers	Workers	`workers.csv`

Run Organizations → Positions → Workers (or on the schedule Atlassian expects).

Monitoring & troubleshooting

OnLink job logs (Bulk Jobs / import log UI) show messages like:

fetched N record(s) from source
uploaded workers.csv with N row(s)
Errors: missing mapping, zero records, S3 PUT failure, invalid required fields

Common errors

Error	Likely cause

Error	Likely cause
`required field(s) not mapped`	Mapping missing a required Teamwork Graph field
`source returned no records`	Wrong `data_path`, filter, or empty HR API response
`all N record(s) were dropped`	Required fields empty or wrong format in source data
`target connection must use AWS Signature V4`	Target is not an S3 Anysource connection
`S3 upload failed with status 403`	IAM permissions or bucket URL incorrect

Example: Anysource HR API → Workers

Source: Anysource connection MyHR API (GET employees)

Mapping:

config:data_path=employees
map:emp_id=workerId
map:created_date=createdAt
map:full_name=name
map:active_flag=isActive
map:email=emailAddress
map:hire_date=hireDate
map:job_title=title

Target: S3 connection Teamwork Graph Bucket

Result: workers.csv uploaded to the bucket; Atlassian HRIS S3 connector ingests it on its next sync.

Teamwork Graph from Any HR Source