| name | jira |
| description | Read-only access to WATER project Jira tickets for task context |
Jira skill
Setup
The scripts require three environment variables:
JIRA_USER — the email address you use to log in to Jira
JIRA_TOKEN — a personal API token (create one here)
JIRA_BASE_URL — the organisation's Jira base URL, e.g. https://eaflood.atlassian.net (no trailing slash)
Run bash .agents/skills/tools/jira/auth.sh to verify the credentials are working before using the skill.
Constraints
This skill is read-only. You may only fetch ticket information. You must not:
- Create tickets
- Update tickets (fields, status, assignee, etc.)
- Transition tickets between statuses
- Add or edit comments
- Upload attachments
If asked to do any of the above, refuse and explain that Jira access is read-only.
Fetching a ticket
Run the ticket script to retrieve and display ticket information:
bash .agents/skills/tools/jira/ticket.sh <NUMBER>
Where <NUMBER> is either the full ticket ID (WATER-4904) or just the number (4904). The script always fetches from the WATER project.
Using ticket context
Once fetched, treat the ticket output as authoritative context for the current task:
- The summary and description define what needs to be done
- The comments may contain additional decisions, constraints, or links to supporting material
- The status and parent give placement within the wider body of work
- Do not infer intent beyond what is written — if the ticket is ambiguous, say so