Initial commit

2025-11-30 08:46:33 +08:00
commit cc6be606c5
6 changed files with 275 additions and 0 deletions
--- a/skills/jira/SKILL.md
+++ b/skills/jira/SKILL.md
@@ -0,0 +1,103 @@
+---
+name: jira
+description: Operate Atlassian Jira through the official jira CLI; trigger this skill to list, inspect, create, update, triage, or report on Jira work using deterministic shell commands aligned with Jira MCP workflows.
+version: 0.1.0
+---
+
+# Jira Issue Management
+
+**Use this skill whenever Jira work is requested.**
+
+## Setup
+
+- Run `./scripts/check-environment.sh`. The skill halts with explicit instructions when `jira` is missing or not authenticated (`jira init` required once).
+- Ensure the `jira` CLI remains in PATH after the check.
+- Use upstream flags like `--project`, `--board`, etc., for context as needed; no extra configuration lives in the skill.
+
+## Command Pattern
+
+```bash
+jira <command> <subcommand> [arguments] [flags]
+```
+
+Primary commands: `issue`, `project`, `board`, `sprint`, `epic`, `release`, `open`, `me`, `serverinfo`.
+
+Keep outputs deterministic:
+
+- Run `./scripts/check-environment.sh` at the start of every session; abort work if it reports an error.
+- Add `--plain` plus `--columns ...` for TSV tables that stream cleanly.
+- Use `--raw` (alias for JSON) when structured data is required.
+- Provide explicit ranges via `--paginate start:limit` to cap interactive views.
+
+## Core Workflows
+
+- **Projects** – enumerate available projects  
+  ```bash
+  jira project list --plain --columns key,name,type
+  ```
+
+- **Issues**  
+  - Search / filter:
+    ```bash
+    jira issue list --status Done --assignee "$(jira me)" --plain --columns key,summary,status,assignee
+    jira issue search 'project = PROJ AND status = "In Progress"' --plain --columns key,summary,status
+    jira issue list --raw --jql 'label = backend' --paginate 0:50
+    ```
+  - View details: `jira issue view PROJ-123 --plain --comments 5`
+  - Create:
+    ```bash
+    jira issue create \
+      --type Bug \
+      --summary "API returns 500" \
+      --description-file docs/bug.md \
+      --priority High \
+      --assignee "$(jira me)" \
+      --label backend --label urgent
+    ```
+  - Update:
+    ```bash
+    jira issue edit PROJ-123 --summary "Refine API contract" --priority Medium
+    jira issue move PROJ-123 "In Review" --comment "Ready for QA"
+    jira issue assign PROJ-123 user@example.com
+    ```
+  - Comment / worklog:
+    ```bash
+    jira issue comment add PROJ-123 "Investigating..." --no-input
+    jira issue worklog add PROJ-123 "1h 30m" --comment "Debugging"
+    ```
+  - Linking: `jira issue link PROJ-1 PROJ-2 "Blocks"` (remove with `jira issue unlink ...`).
+
+- **Epics**  
+  - Create: `jira epic create --name "Platform Stability" --summary "Q3 initiative"`  
+  - List: `jira epic list --plain --columns key,summary,status`  
+  - Attach issues: `jira epic add EPIC-1 PROJ-123 PROJ-456`
+
+- **Boards & Sprints**  
+  - Boards: `jira board list --plain --columns id,name,type`
+  - Sprint overview: `jira sprint list --plain --columns id,name,state,start,end`
+  - Sprint issues: `jira sprint list <SPRINT_ID> --plain --columns key,summary,status,assignee`
+  - Move issues: `jira sprint add <SPRINT_ID> PROJ-123 PROJ-456`
+  - Close sprint: `jira sprint close <SPRINT_ID>`
+
+- **Releases (Versions)** – `jira release list --plain --columns name,start,end,state`
+
+- **Open in browser** – `jira open PROJ-123 --no-browser` (omit flag to launch UI).
+
+- **User / Instance context** – `jira me`, `jira serverinfo`, `jira version`.
+
+## Output Strategies
+
+- Issues default to `key,summary,status,assignee`; sprints to `id,name,state,start,end`. Override with `--columns`.
+- `--raw` returns JSON for downstream processing (`jq`, scripting).
+- Combine commands with standard Unix tooling:
+  ```bash
+  jira issue list --plain --columns key,status,assignee | rg "In Progress"
+  jira issue view PROJ-123 --plain --comments 10 | tee PROJ-123.txt
+  ```
+
+## Reference Material
+
+- `references/mcp-parity.md` contains a Jira MCP → CLI mapping, including attachments, watchers, and worklogs.
+- For authentication or environment issues
+    - re-run `jira init` and review `~/.config/.jira/.config.yml`.
+    - check that JIRA_API_TOKEN is set or `~/.netrc` contains proper jira setup