capp-get-jira-info

name	capp-get-jira-info
description	Centralised CAPP Jira metadata — cloud ID, field IDs, transition IDs, issue type IDs, link types. Invoke before any Jira MCP call to skip redundant discovery API calls and ensure consistent field references.

CaPP Jira Reference Data

Stability Classification

Level	Meaning	Examples
Stable	Changes only via admin action; safe to hardcode	Cloud ID, project ID/key, field IDs, issue type IDs, link type IDs
Workflow-dependent	Changes if the Jira workflow is edited	Transition IDs, status IDs, status category mappings
Dynamic	Changes frequently; always query live	Sprint IDs, user account IDs, board IDs

---

Connection

Key	Value
Cloud ID	3fa693cf-5995-4779-a4db-6af578c534d3
Project key	CAPP
Project ID	14798

Pass cloudId: "3fa693cf-5995-4779-a4db-6af578c534d3" to every Atlassian MCP tool call.

---

Issue Types

Type	ID	Subtask?	Notes
Task	21348	No	Default for general work
Bug	21349	No
Story	21350	No	Only type with Story Points
Epic	21351	No
Subtask	21352	Yes	Created under a parent issue
Spike	25025	No
Unplanned Work	25026	No
Design Task	30779	No
Retro Action	32181	No

When creating issues pass issueTypeName (e.g. "Task") — the MCP tool resolves the name. Use the numeric ID only when an API requires it explicitly.

---

Custom Fields

Field	ID	Available on	Format notes
Dev Notes	customfield_26991	Task, Bug, Story, Epic, Spike, Unplanned Work, Design Task — NOT Subtask	Rich text. Write with contentFormat: "markdown". Read with responseContentFormat: "markdown"
Story Points	customfield_10016	Story only	Number
Sprint	customfield_10020	Most types	Dynamic — always query live
Flagged	customfield_10021	Most types	Option ID 10019 = Impediment

Field Availability Matrix

Field	Task	Bug	Story	Subtask	Epic	Spike
Dev Notes	✅	✅	✅	❌	✅	✅
Story Points	❌	❌	✅	❌	❌	❌
Sprint	✅	✅	✅	✅	❌	✅

---

Workflow & Transitions

Status Map

Status	Category
To Do	To Do
In Progress	In Progress
In Review	In Progress
In Test	In Progress
Ready for Release	Done
Done	Done

Transition Map

From → To	Transition Name	ID
To Do → In Progress	Begin Work	2
In Progress → In Review	In Review	3
In Review → In Test	In Test	4
In Test → Ready for Release	Ready for Release	5
In Progress → Done	Done	6
In Test → Done	Done	6
In Test → In Progress	Failed Testing	7
In Review → In Progress	Failed Review	8
Any → Won't Do	Won't Do	10
In Progress → To Do	ToDo	11
Done → To Do	ReOpen	12
Done → In Progress	In Progress	14

Transition Validation Rule

Always confirm the issue's current status matches the expected "From" status before using a hardcoded transition ID. Jira transition IDs are only valid from specific statuses. If the current status does not match, consult this map to find the correct transition or fall back to live discovery.

---

Link Types

Name	ID	Inward description	Outward description
Blocks	10000	is blocked by	blocks
Cloners	10001	is cloned by	clones
Duplicate	10002	is duplicated by	duplicates
Relates	10003	relates to	relates to

When creating links with createIssueLink, pass the name (e.g. type: "Blocks"). The inward issue is the blocker; the outward issue is the blocked task.

---

Content Format Guidance

Operation	Parameter	Value
Writing description / Dev Notes / comments	contentFormat	"markdown"
Reading issue fields	responseContentFormat	"markdown"
Creating issues	contentFormat	"markdown"

Always pass these parameters explicitly — the MCP default is ADF, which is harder to read and write.

---

Fallback Protocol

If a hardcoded value fails (e.g. transition returns an error, field is not found):

1. Try hardcoded value from this skill first
2. Live discovery once — query the relevant Jira metadata endpoint (transitions, fields, issue types)
3. Continue if unambiguous — if discovery returns the expected value under a different ID, use it and report the mismatch
4. Report mismatch — note the discrepancy so this skill can be updated
5. Ask the user if ambiguous — if discovery returns multiple candidates or no match, do not guess

This protocol avoids both blind trust in stale IDs and unnecessary API calls on every invocation.

---

Dynamic Values (Always Query Live)

These values change too frequently to hardcode:

• Sprint ID — query via customfield_10020 on an existing issue, or list sprints from the board
• User account IDs — use lookupJiraAccountId with search string
• Board ID — query if needed for sprint operations
• Epic link — varies by Jira configuration; check field metadata if linking to epics

---

How Skills Should Use This Data

1. Invoke this skill at the start of any workflow that calls Atlassian MCP tools
2. Use hardcoded values directly for stable and workflow-dependent fields — do not re-discover cloud ID, project ID, field IDs, transition IDs, or link types on every run
3. Follow the transition validation rule — check current status before using a transition ID
4. Follow the fallback protocol if any hardcoded value produces an error
5. Always pass contentFormat/responseContentFormat as specified in the content format table
6. Check the field availability matrix before writing to Dev Notes or Story Points — these are not available on all issue types