From honeycomb
Honeycomb CLI and MCP server usage — auth, queries, datasets, SLOs, boards, triggers, markers, columns, and raw API requests for observability workflows
npx claudepluginhub bendrucker/claude --plugin honeycombThis skill uses the workspace's default tool permissions.
Three key types, stored in the OS keyring per profile:
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).
Generates original PNG/PDF visual art via design philosophy manifestos for posters, graphics, and static designs on user request.
Three key types, stored in the OS keyring per profile:
| Type | Used For |
|---|---|
config | Management API (boards, SLOs, triggers, columns, queries) |
ingest | Sending events (X-Honeycomb-Team header) |
management | Same as config, kept separate for granular access |
honeycomb auth login # interactive key setup
honeycomb auth status # verify stored key
honeycomb auth status --offline # check keyring without API call
Requires a config key with query permissions.
# from a file
honeycomb query run --dataset my-dataset --file query.json
# from stdin
echo '{"calculations":[{"op":"COUNT"}],"time_range":3600}' | honeycomb query run --dataset my-dataset --file -
# list/view saved queries
honeycomb query annotation list --dataset my-dataset
honeycomb query annotation view <id> --dataset my-dataset
honeycomb api makes authenticated requests to any Honeycomb API endpoint. Useful for endpoints without dedicated subcommands (events, query results, service maps).
honeycomb api /1/events/my-dataset -X POST --input events.json --key-type ingest
honeycomb api /1/query_results/my-dataset/abc123 # poll query results
honeycomb api /1/maps/dependencies/requests -q '.results'
honeycomb api /2/teams -q '.[].name' --paginate # v2 with jq + pagination
Flags: -X method, -f string fields, -F typed fields, -H headers, -q jq filter, --paginate, --input body file, --key-type override, --raw skip JSON:API unwrapping (v2 paths).
| Flag | Purpose |
|---|---|
--format json|table | Output format (default: table in TTY, json in CI/agent) |
--profile | Configuration profile |
--no-interactive | Disable prompts |
--api-url | Override Honeycomb API URL |
list, get, create, update, delete subcommands--dataset for dataset-scoped resources (columns, markers, queries), --environment where applicableCLAUDE_CODE is set, the CLI forces --no-interactive and defaults to --format jsonauth, query, dataset, board, column, marker, slo, trigger, environment, key, recipient, mcp, apimcp SubcommandThis plugin configures the Honeycomb MCP server (.mcp.json) for direct tool access via OAuth. Use its tools (run_query, get_dataset_columns, find_columns) when you need query results in context.
The CLI's mcp subcommand is a separate MCP client that authenticates with a config key (Bearer token). Use it when you need to write MCP results to disk:
honeycomb mcp call run_query -f dataset=prod -q '.content[].text' > results.json
honeycomb mcp tools # list available MCP tools