From atlassian
Access Confluence data via the Atlassian CLI (`acli`). Use when the user asks about Confluence pages, spaces, blog posts, or otherwise needs to interact with Confluence.
How this skill is triggered — by the user, by Claude, or both
Slash command
/atlassian:confluenceThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Prefer the Atlassian CLI tool (`acli`) for interacting with Confluence. For anything `acli` does not cover (notably full-text search), fall back to the authenticated `curl` wrapper at `scripts/atlassian-curl.sh`.
Prefer the Atlassian CLI tool (acli) for interacting with Confluence. For anything acli does not cover (notably full-text search), fall back to the authenticated curl wrapper at scripts/atlassian-curl.sh.
acliacli confluence --help shows the available Confluence subcommands (currently auth, blog, page, and space).--help; for example, to view a Confluence page by ID:
acli confluence page view --helpscripts/atlassian-curl.sh (for endpoints acli does not expose)The wrapper takes a URL path plus arbitrary curl options and prints the JSON response. Credentials are read from the environment by the script itself; you do not need to pass any auth-related arguments, and you do not need to read or modify the script. Treat it as an opaque tool.
If the script reports that a required environment variable is unset, surface that to the user and stop — do not try to work around it.
In the examples below, $SKILL_DIR is the absolute path to the directory containing the SKILL.md file that defines this skill. Always invoke the wrapper as $SKILL_DIR/scripts/atlassian-curl.sh (with $SKILL_DIR expanded to its absolute value before running the command); do not invoke it as a bare relative path, since the current working directory is not guaranteed to be the skill directory.
Usage shape:
$SKILL_DIR/scripts/atlassian-curl.sh <path> [curl options...]
Notes:
<path> begins with / (e.g. /wiki/rest/api/search) and is appended to the configured Atlassian site.Accept: application/json. Do not pass -v / --verbose (it would echo the Authorization header to stderr)./rest/api/3/...); prefer acli jira first and only fall back to the wrapper for things acli lacks.The Confluence Cloud search endpoint is GET /wiki/rest/api/search and accepts a CQL expression in the cql query parameter. Build the CQL from the user's request:
text ~ "some phrase"title ~ "runbook"space = "ENG"type = page (or blogpost, attachment, space, user)AND / OR, parentheses, NOT.Use --data-urlencode so you do not have to hand-encode the CQL:
$SKILL_DIR/scripts/atlassian-curl.sh /wiki/rest/api/search \
--get \
--data-urlencode 'cql=type = page AND text ~ "incident response"' \
--data-urlencode 'limit=10'
The response is JSON with a results array. Each entry has title, excerpt, url (relative — prefix with https://<site>/wiki for a clickable link), and a nested content object with the page id, type, and space. Pipe through jq to summarise, e.g.:
$SKILL_DIR/scripts/atlassian-curl.sh /wiki/rest/api/search \
--get \
--data-urlencode 'cql=space = "ENG" AND type = page AND text ~ "deploy"' \
--data-urlencode 'limit=10' \
| jq '.results[] | {title, id: .content.id, excerpt}'
For pagination, pass --data-urlencode 'start=N'; the response includes start, limit, size, and a _links.next field when more results are available.
Searches MemPalace before answering questions about past work, people, projects, or prior decisions. Returns verbatim stored content instead of guessing from model memory.
Guides Payload CMS config (payload.config.ts), collections, fields, hooks, access control, APIs. Debugs validation errors, security, relationships, queries, transactions, hook behavior.
Implements vector databases with Pinecone, Weaviate, Qdrant, Milvus, pgvector for semantic search, RAG, recommendations, and similarity systems. Optimizes embeddings, indexing, and hybrid search.
npx claudepluginhub wincent/wincent-agent-plugins --plugin atlassian