Official skill collection for Cleo — the open-source Slack bot framework.
| Skill | Type | Tools | Credentials |
|---|---|---|---|
| Jira | Python | jira_search, jira_get_issue, jira_create_issue, jira_update_issue, jira_transition_issue, jira_get_sprints |
atlassian |
| Confluence | Python | confluence_search, confluence_read_page, confluence_create_page, confluence_update_page |
atlassian |
| Slack Search | Python | slack_search_messages, slack_read_thread |
slack |
For Jira and Confluence, Atlassian has an official MCP server. You can use it with Cleo by adding to cleo.yaml:
skills:
- type: mcp
name: atlassian
transport: sse
url: https://mcp.atlassian.com/v1/sseWe ship Python skills as well because:
- Per-user credentials — Cleo passes each user's own token to the skill. MCP servers typically use a single shared credential, which means everyone appears as the same user.
- No external dependency — Python skills run in-process. No subprocess, no npx, no Node.js required.
- Customizable — fork and modify tool definitions, add fields, change behavior.
Use whichever fits your setup. Both work. If your org is fine with a shared service account for Jira/Confluence, the MCP server is simpler. If you need per-user attribution, use these Python skills.
For Slack Search, there is no MCP server that supports user tokens (xoxp-), so the Python skill is the only option for per-user search.
pip install cleo-skillsOnce installed, Cleo auto-discovers the skills on startup. No configuration needed.
Users connect their credentials via DM:
/connect atlassian sarah@company.com YOUR_ATLASSIAN_API_TOKEN
/connect slack YOUR_XOXP_TOKEN
Tools:
jira_search— Search with JQL (e.g.project = PLAT AND status = "In Progress")jira_get_issue— Full details for a single issue (description, comments, links)jira_create_issue— Create issues with project, type, summary, description, priority, assigneejira_update_issue— Update fields or add commentsjira_transition_issue— Change status (e.g. move to "Done")jira_get_sprints— List sprints for a board (active, future, closed)
API: Jira Cloud REST API v3
Tools:
confluence_search— Search with CQL (e.g.text ~ "auth redesign" AND space = PLAT)confluence_read_page— Read full page content (by ID or title)confluence_create_page— Create page in a space (with optional parent page)confluence_update_page— Update existing page content
API: Confluence Cloud REST API
Note: Jira and Confluence share the same Atlassian credential — one /connect atlassian covers both.
Tools:
slack_search_messages— Search across all channels user has access toslack_read_thread— Read full thread replies
API: Slack Web API (requires user token xoxp- for per-user search scope)
See the Cleo README for the full skill authoring guide.
MIT