jira-recipes

name	jira-recipes
description	Jira integration recipes for Workato. Enables AI agents to generate valid recipe JSON for Jira operations including searching issues by JQL, creating and updating issues, and querying sprint data.
license	MIT
metadata	{"author":"Workato","version":"1.0.0"}

Jira Recipes Skill - Agent Instructions

⚠️ DEPENDENCY: Load the workato-recipes base skill first if not already loaded. This skill requires the base Workato knowledge for triggers, control flow, datapills, formulas, and recipe structure.

This skill provides Jira-specific knowledge for generating Workato recipes. It extends the workato-recipes base skill and focuses on Jira-specific patterns.

---

CRITICAL: Pre-Generation Checklist

For EXISTING projects:

1. Read existing Jira .recipe.json files to understand local patterns

For GREENFIELD projects:

1. Use skill templates - see templates/search-issues-by-jql.json as reference
2. Use descriptive UUIDs - e.g., jira-search-sprint-003, return-success-jira-004

ALWAYS:

1. Ask for connection name - exact name of Jira connection in Workato
2. Confirm project key - the Jira project key (e.g., ENG, MOB, PROJ)
3. Use API endpoint trigger for testability via curl
4. Use descriptive UUIDs - never copy random hex UUIDs from existing recipes
5. Remember action name casing - search_issues_by_JQL is case-sensitive (uppercase JQL)

---

Table of Contents

1. When to Use This Skill
2. Jira Config Requirements
3. Native Connector Guidance
4. JQL Query Syntax
5. Jira Datapill Paths
6. Jira Patterns
7. Extended Input Schema Rules
8. Common Errors
9. Validation

---

When to Use This Skill

Use this skill when building Workato recipes that:

• Search Jira issues using JQL queries
• Query sprint status and issue breakdowns
• Create or update Jira issues
• Retrieve issue details by ID or key

Prerequisites:

• workato-recipes base skill loaded
• Workato workspace with Jira connection configured
• Understanding of the target Jira project structure (project keys, sprint names)

---

Jira Config Requirements

Every Jira recipe requires the jira provider in the config section:

{
  "keyword": "application",
  "provider": "jira",
  "skip_validation": false,
  "account_id": {
    "zip_name": "connectors/jira.connection.json",
    "name": "My Jira Connection",
    "folder": "connectors"
  }
}

Combined with API endpoint trigger:

"config": [
  {
    "keyword": "application",
    "provider": "workato_api_platform",
    "skip_validation": false,
    "account_id": null
  },
  {
    "keyword": "application",
    "provider": "jira",
    "skip_validation": false,
    "account_id": {
      "zip_name": "connectors/jira.connection.json",
      "name": "My Jira Connection",
      "folder": "connectors"
    }
  }
]

---

Native Connector Guidance

The Jira connector provides 17 native actions and 11 triggers. See lint-rules.json for the authoritative list of valid action and trigger names.

Choosing the Right Trigger

• Generic webhook: new_event — fires on any configured Jira event (issue created, updated, etc.).
• Issue polling: new_issue, updated_issue — poll for new or updated issues.
• Batch polling: new_issue_batch, updated_issue_batch — poll for batches of new/updated issues.
• Bulk: issue_created_bulk, issue_created_or_updated_bulk — for high-volume processing.
• Specific webhooks: updated_comment_webhook, updated_issue_webhook, updated_worklog_webhook — real-time triggers for specific change types.
• Deletion: deleted_object — triggers on object deletion.

Choosing the Right Action

Issue CRUD:

• create_issue — Create a new issue. Key inputs: project_key, issue_type, summary. See detail below.
• get_issue — Get full issue details by ID or key. See detail below.
• update_issue — Update issue fields. Requires issue_id_or_key. See detail below.
• update_issue_status — Transition issue to a different workflow state.
• assign_issue — Assign an issue to a user.

Search:

• search_issues_by_JQL — Primary search action. Full JQL flexibility, battle-tested. Name is case-sensitive (uppercase JQL). See detail below.
• search_issues — Simpler field-filter search without JQL. Less flexible.

Comments: create_comment, update_comment, get_issue_comments

Users: find_user (by query string), search_assignable_users (users assignable to an issue), create_user

Attachments: upload_attachment, get_attachment

Metadata: get_changelog (issue change history), get_object_schema

Uncovered operations: Use __adhoc_http_action for Jira operations not covered by native actions: workflow transitions, sprint/board management, project listing, custom field management, tempo/time tracking, and advanced operations.

---

search_issues_by_JQL Detail

Use when: Finding issues using JQL queries — sprints, filters, project scoping.

CRITICAL: The action name is case-sensitive. It MUST be search_issues_by_JQL (uppercase JQL).

{
  "provider": "jira",
  "name": "search_issues_by_JQL",
  "as": "jira_search",
  "keyword": "action",
  "input": {
    "jql": "='project = \"' + _dp('{...project_key...}') + '\" AND sprint in openSprints()'"
  },
  "extended_input_schema": [],
  "extended_output_schema": [
    {
      "label": "Issues",
      "name": "issues",
      "type": "array",
      "of": "object",
      "properties": [
        { "control_type": "text", "label": "Key", "name": "key", "type": "string" },
        { "control_type": "text", "label": "Summary", "name": "summary", "type": "string" },
        { "control_type": "text", "label": "Status", "name": "status", "type": "string" },
        { "control_type": "text", "label": "Assignee", "name": "assignee", "type": "string" },
        { "control_type": "number", "label": "Story Points", "name": "story_points", "type": "number" }
      ]
    }
  ]
}

Key rules:

• jql is the native field name (NOT query) — maps to "JQL query string" in the Workato UI
• extended_input_schema MUST be [] — jql is a connector internal; including it in EIS creates a duplicate field
• JQL string is built using = formula mode (bare _dp(), no #{} wrapper)
• extended_output_schema defines the issue fields you want to access via datapills

create_issue Detail

{
  "provider": "jira",
  "name": "create_issue",
  "as": "create_jira_issue",
  "keyword": "action",
  "input": {
    "project_key": "ENG",
    "issue_type": "Story",
    "summary": "#{summary_datapill}",
    "description": "#{description_datapill}"
  }
}

Expected fields: project_key, issue_type, summary, description, priority, assignee, labels

update_issue Detail

{
  "provider": "jira",
  "name": "update_issue",
  "as": "update_jira_issue",
  "keyword": "action",
  "input": {
    "issue_id_or_key": "ENG-123",
    "summary": "#{summary_datapill}"
  }
}

get_issue Detail

{
  "provider": "jira",
  "name": "get_issue",
  "as": "get_jira_issue",
  "keyword": "action",
  "input": {
    "issue_id_or_key": "ENG-123"
  }
}

---

JQL Query Syntax

JQL queries in Workato are built using = formula mode. This means bare _dp() calls — no #{} interpolation.

Dynamic JQL Construction

The jql input field uses formula mode (= prefix) to concatenate strings and datapills:

"jql": "='project = \"' + _dp('{...project_key...}') + '\" AND sprint in openSprints()'"

Breaking this down:

• = — enters formula mode
• 'project = \"' — literal string with escaped double quotes for JQL values
• + _dp('{...}') + — concatenates the datapill value
• '\" AND sprint in openSprints()' — more literal JQL

Conditional Clauses with .present?

Use the ternary pattern to add optional clauses:

"jql": "='project = \"' + _dp('{...project_key...}') + '\"' + (_dp('{...sprint_name...}').present? ? ' AND sprint = \"' + _dp('{...sprint_name...}') + '\"' : ' AND sprint in openSprints()')"

This pattern:

• Checks if sprint_name was provided
• If yes: filters by the specific sprint name
• If no: defaults to all open sprints

IMPORTANT: No outer () wrapping the entire formula. The = is followed directly by the expression. Wrapping breaks condition LHS parsing in child actions.

See: patterns/jql-query-syntax.md for complete reference.

---

Jira Datapill Paths

No Body Wrapper

Jira actions do NOT use the ["body"] wrapper in datapill paths. This is consistent with other native connectors (Salesforce, Stripe).

// CORRECT for Jira
"path": ["issues"]
"path": ["issues", {"path_element_type": "current_item"}, "key"]

// WRONG for Jira — Do NOT use
"path": ["body", "issues"]

Jira Datapill Examples

Issues array (for .to_json serialization):

"#{_dp('{\"pill_type\":\"output\",\"provider\":\"jira\",\"line\":\"jira_search\",\"path\":[\"issues\"]}')}"

Issue count (using path_element_type: "size"):

"=_dp('{\"pill_type\":\"output\",\"provider\":\"jira\",\"line\":\"jira_search\",\"path\":[\"issues\",{\"path_element_type\":\"size\"}]}')"

Issues array as JSON string (for return_response):

"=_dp('{\"pill_type\":\"output\",\"provider\":\"jira\",\"line\":\"jira_search\",\"path\":[\"issues\"]}').to_json"

Individual issue field (inside foreach):

"path": ["issues", {"path_element_type": "current_item"}, "key"]
"path": ["issues", {"path_element_type": "current_item"}, "summary"]
"path": ["issues", {"path_element_type": "current_item"}, "status"]

---

Jira Patterns

1. Search + Serialize Pattern

Search for issues and return as JSON string with count:

// Step: Jira search
{
  "provider": "jira",
  "name": "search_issues_by_JQL",
  "as": "jira_search",
  "input": {
    "jql": "='project = \"ENG\" AND sprint in openSprints()'"
  },
  "extended_input_schema": [],
  "extended_output_schema": [...]
}

// Step: Return response
{
  "provider": "workato_api_platform",
  "name": "return_response",
  "input": {
    "http_status_code": "200",
    "response": {
      "issues_json": "=_dp('{...path:[\"issues\"]}').to_json",
      "total_issues": "=_dp('{...path:[\"issues\",{\"path_element_type\":\"size\"}]}')"
    }
  }
}

Key: Use .to_json to serialize the issues array into a string for return. Use path_element_type: "size" for the count — NOT .size formula method.

2. Input Validation Pattern

Always validate required inputs before making Jira API calls:

{
  "keyword": "if",
  "input": {
    "conditions": [{
      "operand": "present",
      "lhs": "#{_dp('{...path:[\"request\",\"project_key\"]}')}",
      "uuid": "cond-project-key-001"
    }]
  },
  "block": [
    // Jira search goes here
  ]
}

3. Try/Catch for Jira Errors

Wrap Jira actions in try/catch to handle API failures:

{
  "keyword": "try",
  "block": [
    // Jira actions...
  ]
},
{
  "keyword": "catch",
  "provider": null,
  "as": "catch_jira_error",
  "input": { "max_retry_count": "0", "retry_interval": "2" },
  "block": [
    {
      "provider": "workato_api_platform",
      "name": "return_response",
      "input": {
        "http_status_code": "500",
        "response": {
          "error_message": "#{_dp('{\"pill_type\":\"output\",\"provider\":\"catch\",\"line\":\"catch_jira_error\",\"path\":[\"message\"]}')}"
        }
      }
    }
  ]
}

Key: Catch datapills use "provider":"catch" — NOT "provider":null.

---

Extended Input Schema Rules

CRITICAL: jql Is a Connector Internal

The jql field is Jira's native parameter for the search_issues_by_JQL action. It must NEVER appear in extended_input_schema.

Correct:

{
  "input": {
    "jql": "='project = \"ENG\"'"
  },
  "extended_input_schema": []
}

Wrong — causes duplicate field in UI:

{
  "input": {
    "jql": "='project = \"ENG\"'"
  },
  "extended_input_schema": [
    { "name": "jql", "type": "string", ... }
  ]
}

General Rule

Empty extended_input_schema: [] is correct for search_issues_by_JQL because the only input (jql) is a connector internal. If future actions require user-facing fields that are NOT connector internals, those would go in EIS — but for search, [] is the right answer.

---

Common Errors

Error	Cause	Solution
"Select an app and action" in UI	Action name is wrong or case-incorrect	Use exactly search_issues_by_JQL (uppercase JQL)
Duplicate JQL field in UI	jql included in extended_input_schema	Remove jql from EIS; set extended_input_schema: []
"Field not recognized"	Using query instead of jql	The native field name is jql, not query
JQL parse error	Missing double quotes around values in JQL string	Use \"value\" inside formula mode strings
Empty results	JQL formula not in = mode	Prefix JQL value with = for formula mode
Datapill not found	["body"] wrapper in path	Jira is a native connector — no ["body"] wrapper

---

Validation

See validation-checklist.md for Jira-specific validation, which references the base checklist in workato-recipes/validation-checklist.md.

---

Templates

See templates/ directory:

• search-issues-by-jql.json - Search issues by JQL with conditional sprint filter, try/catch, multi-status responses

---

References

• Base Skill: workato-recipes - Recipe structure, triggers, control flow, formulas
• Templates: templates/ directory
• Patterns: patterns/ directory
• Jira Cloud REST API: https://developer.atlassian.com/cloud/jira/platform/rest/v3/
• JQL Reference: https://support.atlassian.com/jira-service-management-cloud/docs/use-advanced-search-with-jira-query-language-jql/