kangaplanner-api
Read and write KangaPlanner issues (bugs, features, tasks) over its REST API at api.kangasuite.com with a kp_ bearer token. Use whenever the user mentions KangaPlanner, planner.kangasuite.com, a Kanga
SKILL.mdusers/kyandesutter/claude/skills/kangaplanner-api/SKILL.md
GitHubKangaPlanner API
Base URL: https://api.kangasuite.com/api/issues (the web app is
planner.kangasuite.com; settings > API tokens is where tokens are minted).
Auth: Authorization: Bearer kp_…. The token lives at
~/.claude/secrets/kangaplanner-api-token (chmod 600). Read it into a shell
variable, never print it or write it into a file:
T=$(cat ~/.claude/secrets/kangaplanner-api-token)
B=https://api.kangasuite.com/api/issues
curl -s "$B" -H "Authorization: Bearer $T" | jq '.issues | length'
The token acts as its owning user, so it sees every company and space that user
can see. Scopes are issues:read (GET) and issues:write (everything else).
Errors come back as {"error": "..."} with 400/401/403/404/409/429; a 400 body is
a stringified Zod issue list.
Endpoints
| Method | Path | Body / query | Returns |
|---|---|---|---|
| GET | / |
?spaceId= &includeArchived=true &assignedToMe=true &reportedByMe=true |
{issues: [...]} |
| GET | /:id |
{issue} with replies, assignees, followers, attachments |
|
| POST | / |
create body below | {issue} (201) |
| PATCH | /:id |
update body below | {issue} |
| POST | /:id/replies |
{body, audience?} |
{reply} (201) |
| GET | /:id/attachments/:aid |
raw bytes | |
| POST | /:id/convert |
{toType} |
{issue}, ref changes |
:id is always the issue's UUID. Refs like B-BUG-69 are rejected, so resolve
a ref by listing and filtering:
curl -s "$B" -H "Authorization: Bearer $T" | jq -r '.issues[] | select(.ref=="B-BUG-69") | .id'
The list returns several hundred rows with description cut to a 300-char
snippet. Always narrow it with jq ({id, ref, subject, status, priority, space: .space.name, assignees: [.assignees[].name]}) rather than dumping it.
There is no spaces endpoint: get space names and ids from the list rows
([.issues[] | {spaceId, space: .space.name}] | unique).
Bodies
Enums: type bug|feature|task; priority low|medium|high|critical; status
open|in_progress|resolved|closed; blockedReason waiting_reporter|on_hold
(only while in_progress); closeReason
fixed|duplicate|not_a_bug|by_design|wont_fix|cant_reproduce (only while
closed). Dates are YYYY-MM-DD; start may not be after due.
Create (required: spaceId, subject ≤255, description ≤50k, markdown):
{"spaceId": "<uuid>", "subject": "…", "description": "…", "type": "bug",
"priority": "medium", "startDate": "2026-10-21", "dueDate": "2026-10-31",
"assigneeIds": ["<uuid>"], "followerIds": ["<uuid>"]}
Update: any subset of subject, description, priority, status,
startDate, dueDate, blockedReason, closeReason, reporterEmail. Omit to
leave unchanged, null to clear a date, reason or email.
Reply: {"body": "…", "audience": "internal"}. Audience defaults to internal
(team-only). client emails the portal client and end_user emails the
external reporter, so only use those when the user asked to message them.
Convert: {"toType": "feature"} renumbers the issue, so re-read the ref.
Not available over this API
Assigning or unassigning after creation, listing users, and listing spaces
directly. The backend repo (~/Developer/KangaPlanner) has an MCP server at
/mcp that adds these, but it is not deployed yet (404 as of 2026-10-20).