npx claudepluginhub twofoldtech-dakota/plugin-hiveThis skill is limited to using the following tools:
You are the Plugin Hive coordinator. Route the user's request to the appropriate sub-command below.
Runs the hive experiment loop for autonomous iteration on shared tasks in hive directories. Use to run experiments, submit results, or join agent swarms.
Queries, manages, and troubleshoots Apache Airflow using the af CLI. Covers listing DAGs, triggering runs, reading task logs, diagnosing failures, debugging DAG import errors, checking connections, variables, pools, and monitoring health. Also routes to sub-skills for writing DAGs, debugging, deploying, and migrating Airflow 2 to 3. Use when user mentions "Airflow", "DAG", "DAG run", "task log", "import error", "parse error", "broken DAG", or asks to "trigger a pipeline", "debug import errors", "check Airflow health", "list connections", "retry a run", or any Airflow operation. Do NOT use for warehouse/SQL analytics on Airflow metadata tables — use analyzing-data instead.
Guides claude-flow multi-agent orchestration: swarm vs hive-mind topology selection, agent deployment, memory configuration, SPARC workflows. Interactive recommendations without auto-execution.
Share bugs, ideas, or general feedback.
You are the Plugin Hive coordinator. Route the user's request to the appropriate sub-command below.
Arguments: $ARGUMENTS
Parse the first word of $ARGUMENTS as the sub-command. If no arguments are given, show the Help section.
| Sub-command | Action |
|---|---|
swarm <blueprint> <task...> | Start a new swarm (see swarm) |
status [query] | Check swarm status (see status) |
list [--status=X] | List swarms (see list) |
stop <number|id> | Stop a swarm (see stop) |
resume <number|id> | Resume a failed swarm (see resume) |
blueprints | List available blueprints (see blueprints) |
install <id> | Install a blueprint (see install) |
uninstall <id> | Uninstall a blueprint (see uninstall) |
beekeeper | Run a health check (see beekeeper) |
cells <number|id> | List cells for a swarm (see cells) |
pollinate [swarm_id] | Trigger a pollination cycle (see pollinate) |
approve <flight_id> | Approve a gated flight (see approve) |
analytics <number|id> | Show swarm performance analytics (see analytics) |
info <blueprint_id> | Show blueprint info including input schema (see info) |
scaffold <id> [--global] | Scaffold a new blueprint directory (see scaffold) |
validate <id> | Validate a blueprint with semantic checks (see validate) |
dryrun <id> [vars...] | Simulate pipeline execution (see dryrun) |
install-remote <url> | Install blueprint from Git repo (see install-remote) |
chain <chain_id> | View chain status (see chain) |
chains [--status=X] | List all chains (see chains) |
snapshot <number|id> | Create a swarm snapshot (see snapshot) |
snapshots <number|id> | List snapshots for a swarm (see snapshots) |
restore <snapshot_id> | Restore swarm to snapshot (see restore) |
trace <flight_id|swarm_id> | View flight/swarm execution traces (see trace) |
notify config [url] [--format=X] | Get/set notification config (see notify config) |
notify test [url] | Send test webhook (see notify test) |
notify history [--status=X] | View webhook delivery history (see notify history) |
notify retry [delivery_id] | Retry failed webhook deliveries (see notify retry) |
usage <number|id> | Show token usage for a swarm (see usage) |
bee-stats [bee_id] | Show bee performance history (see bee-stats) |
progress <flight_id|number> | Show live flight progress pulses (see progress) |
drive [blueprint task|number] | Autonomously drive a swarm (suggest /hive-drive) |
observatory [start|stop|status] | Manage the Observatory dashboard (see observatory) |
queue | Show concurrency queue status (see queue) |
archive <number|id> | Archive a completed/failed swarm (see archive) |
report <number|id> [--format=X] | Generate a swarm report (see report) |
storage | Show database storage status (see storage) |
config [key] [value] | Get/set global configuration (see config) |
replay <number|id> [--task="..."] [--priority=N] [--reset-nectar] | Replay a completed/failed swarm (see replay) |
metrics [--period=30d] | Show fleet-level aggregate metrics (see metrics) |
maintain [--dry-run] | Run data maintenance cleanup (see maintain) |
export <blueprint_id> [output_dir] | Export a blueprint as portable bundle (see export) |
import <path> | Import a blueprint from a bundle file (see import) |
estimate <blueprint_id> | Predict swarm cost/duration (see estimate) |
gates | List pending gated flights (see gates) |
tune <blueprint_id> [--apply] | Analyze and tune bee parameters (see tune) |
nectar set <N> <key> <value> | Set a nectar key on a swarm (see nectar set) |
nectar get <N> [key] | Get nectar values from a swarm (see nectar get) |
history <blueprint_id> | View blueprint version history (see history) |
diff <blueprint_id> [from] [to] | Diff blueprint versions (see diff) |
budget <number|id> [amount] | View/set token budget for a swarm (see budget) |
cache [clear] | View cache stats or clear cache (see cache) |
compare <N> <M> | Compare two swarm runs side-by-side (see compare) |
inject <N> <after_flight_id> <bee_id> <input> | Inject a flight into a running pipeline (see inject) |
skip <flight_id> [reason] | Skip a pending/waiting flight (see skip) |
template save <name> <blueprint_id> | Save a swarm template (see template save) |
template list | List saved templates (see template list) |
template run <name> <task> | Start a swarm from a template (see template run) |
dag <number|id> | Show DAG visualization (see dag) |
subswarm <flight_id> | Check sub-swarm status (see subswarm) |
routing [flight_id|swarm_id|bee_id] | View model routing history (see routing) |
anomalies [--severity=X] [--status=X] | List anomaly alerts (see anomalies) |
anomaly ack <alert_id> | Acknowledge an anomaly (see anomaly ack) |
baselines [blueprint_id] | View performance baselines (see baselines) |
stream status | Check SSE stream status (see stream status) |
nectar shares <number|id> | List nectar shares (see nectar shares) |
nectar resolve <number|id> <source> <keys> | Resolve a nectar ref (see nectar resolve) |
registry search <query> | Search blueprint registry (see registry search) |
registry install <id> | Install from registry (see registry install) |
rate <blueprint_id> <1-5> [comment] | Rate a blueprint (see rate) |
channel create <name> <type> <config> | Create notification channel (see channel create) |
channel list | List notification channels (see channel list) |
channel delete <id> | Delete notification channel (see channel delete) |
route create <channel_id> <pattern> | Create notification route (see route create) |
route list | List notification routes (see route list) |
route delete <id> | Delete notification route (see route delete) |
webhook token create <name> <perms> | Create webhook token (see webhook token create) |
webhook token list | List webhook tokens (see webhook token list) |
webhook token revoke <id> | Revoke webhook token (see webhook token revoke) |
webhook audit [--limit=N] | View webhook audit log (see webhook audit) |
schedule create <name> <bp> <cron> <task> | Create a recurring schedule (see schedule create) |
schedule list [--blueprint=X] | List schedules (see schedule list) |
schedule delete <id> | Delete a schedule (see schedule delete) |
schedule toggle <id> | Enable/disable a schedule (see schedule toggle) |
schedule history <id> | View schedule run history (see schedule history) |
schedule evaluate | Manually evaluate due schedules (see schedule evaluate) |
circuits [--state=X] | List circuit breaker states (see circuits) |
circuit reset <bee_id> | Reset a circuit breaker (see circuit reset) |
dlq [swarm_id] | List dead-lettered flights (see dlq list) |
dlq replay <id> | Replay a dead-lettered flight (see dlq replay) |
dlq purge [id|--swarm=N] | Purge dead letters (see dlq purge) |
test add <bp> <name> | Add a blueprint test case (see test add) |
test list <bp> | List blueprint test cases (see test list) |
test run <bp|test_id> | Run blueprint test(s) (see test run) |
test delete <test_id> | Delete a test case (see test delete) |
health | Show hive health score (see health) |
health history | Show health score history (see health history) |
tag <N> <key>=<value> | Tag a swarm with key=value metadata. Call hive_swarm_tag with the resolved swarm ID, key, and value |
untag <N> <key> | Remove a tag from a swarm. Call hive_swarm_untag with the resolved swarm ID and key |
search <query> [--tag=X] [--from=date] [--to=date] | Search swarms with rich filters. Call hive_swarm_search with parsed filters |
profile create <name> | Create a config profile. Call hive_profile_create with the name |
profile list | List profiles. Call hive_profile_list |
profile activate <name> | Activate a profile. Call hive_profile_activate with the name |
profile delete <name> | Delete a profile. Call hive_profile_delete with the name |
memory <bee_id> | View bee memories. Call hive_bee_memory_recall with the bee_id |
memory clear <bee_id> | Clear bee memories. Call hive_bee_memory_forget with the bee_id |
deps [blueprint_id] | Show blueprint dependencies. Call hive_blueprint_deps |
playbook create <name> | Create an operational playbook (interactive). Call hive_playbook_create |
playbook list | List playbooks. Call hive_playbook_list |
playbook delete <id> | Delete a playbook. Call hive_playbook_delete with the id |
playbook toggle <id> | Enable/disable a playbook. Call hive_playbook_toggle with the id |
playbook history [id] | View playbook execution history. Call hive_playbook_history |
Display this command reference as a formatted table. Include a brief description of Plugin Hive: "Multi-agent swarm orchestration for Claude Code. Deploy specialized bees to autonomously execute development workflows."
Start a new swarm from a blueprint.
swarm is the blueprint ID. Remaining args are the task description.mcp__hive__hive_blueprint_list and ask the user to pick one.--priority=N (1-10, default 5), --schedule-at=ISO8601 (delay start).mcp__hive__hive_swarm_start with the blueprint_id, task, and optional priority/schedule_at.mcp__hive__hive_pollinate with the returned swarm_id to kick off the first work cycle./hive-drive to run this swarm autonomously."Check the status of a swarm.
mcp__hive__hive_swarm_status with it as the query.mcp__hive__hive_cell_list with the resolved swarm ID to show cell progress.mcp__hive__hive_swarm_list to show all swarms.List swarms with optional filters.
--status=X from args if present.mcp__hive__hive_swarm_list with the status filter.#number | blueprint | task (truncated) | status | created.Stop a running swarm.
1, 3) or a swarm ID.mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_swarm_stop with the resolved swarm ID.Resume a failed or paused swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_swarm_resume with the resolved swarm ID.mcp__hive__hive_pollinate with the swarm ID to immediately kick off work.List available and installed blueprints.
mcp__hive__hive_blueprint_list.Install a blueprint.
mcp__hive__hive_blueprint_install with it.mcp__hive__hive_blueprint_list and prompt the user to choose.Uninstall a blueprint.
mcp__hive__hive_blueprint_uninstall with the given ID.Run a health check on the hive.
mcp__hive__hive_beekeeper_check.mcp__hive__hive_beekeeper_status to show recent check history.List cells for a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_cell_list with the resolved swarm ID.index | cell_id | title | status | retries.Trigger a pollination cycle.
mcp__hive__hive_pollinate with it.mcp__hive__hive_pollinate without a swarm_id (polls all swarms).Approve a gated flight to unblock the swarm pipeline.
mcp__hive__hive_swarm_list with status=blocked to find blocked swarms, then show their gated flights.mcp__hive__hive_gate_approve with the flight_id.Show performance analytics for a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_swarm_analytics with the resolved swarm ID.Show detailed blueprint information including input schema.
mcp__hive__hive_blueprint_info with the given blueprint ID.Scaffold a new blueprint directory.
--global is passed, set location to "global"; otherwise use "project" (default).mcp__hive__hive_blueprint_scaffold with the blueprint_id and location.blueprint.yml.Validate a blueprint with extended semantic checks.
mcp__hive__hive_blueprint_validate with the blueprint_id.Simulate pipeline execution without spawning bees.
key=value pairs are parsed as variables.mcp__hive__hive_blueprint_dryrun with the blueprint_id and variables.Install a blueprint from a Git repository URL.
--subdir=X is given, pass it as the subdirectory.mcp__hive__hive_blueprint_install_remote with the url and optional subdirectory.View chain status with parent-child swarm relationships.
mcp__hive__hive_chain_status with the chain_id.List all chains with optional status filter.
--status=X from args if present (active, completed, failed).mcp__hive__hive_chain_list with the status filter.chain_id | root_swarm | status | created.Create a snapshot of a swarm's state.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_snapshot_create with the resolved swarm ID.List snapshots for a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_snapshot_list with the resolved swarm ID.snapshot_id | type | created_at.Restore a swarm to a snapshot state.
mcp__hive__hive_snapshot_restore with the snapshot_id.View execution traces for a flight or swarm.
mcp__hive__hive_flight_trace with flight_id.mcp__hive__hive_flight_trace with swarm_id.Get or set global notification configuration.
mcp__hive__hive_notification_config with url set.--format=X is given (standard/slack/discord), include the format.--events=X,Y is given, include the events array.mcp__hive__hive_notification_config with no params to show current config.Send a test webhook to verify the configuration.
mcp__hive__hive_notification_test.View webhook delivery history.
--status=X from args if present (pending/delivered/failed).mcp__hive__hive_notification_history with the status filter.delivery_id | event_id | url | status | attempts | last_error.Retry failed webhook deliveries.
mcp__hive__hive_notification_retry with it.mcp__hive__hive_notification_retry without delivery_id to retry all failed.Show token usage breakdown for a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_swarm_usage with the resolved swarm ID.Show bee performance history.
feature-dev_worker), call mcp__hive__hive_bee_stats with bee_id.feature-dev), call mcp__hive__hive_bee_stats with blueprint_id.mcp__hive__hive_bee_stats with no params to show all bee stats.bee_id | flights | success_rate | avg_duration | total_tokens.Show live flight progress pulses.
mcp__hive__hive_flight_progress with flight_id.mcp__hive__hive_flight_progress with swarm_id.Manage the Observatory dashboard web UI.
observatory as the action: start, stop, or status.mcp__hive__hive_observatory_status to show current state.start [port] — Call mcp__hive__hive_observatory_start with optional port. Report the URL on success.stop — Call mcp__hive__hive_observatory_stop. Confirm shutdown.status — Call mcp__hive__hive_observatory_status. Report running state, PID, port, and URL.Show concurrency queue status.
mcp__hive__hive_queue_status.Archive a completed, failed, or cancelled swarm to compressed storage.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_swarm_archive with the resolved swarm ID.Generate a structured report for a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.--format=X from args if present (json, markdown, both). Default is "both".mcp__hive__hive_swarm_report with the resolved swarm ID and format.Show database storage status.
mcp__hive__hive_storage_status.Get or set global configuration.
mcp__hive__hive_config with no params to show all config entries with descriptions.mcp__hive__hive_config with just the key to show that entry.mcp__hive__hive_config with key and value to update it.max_concurrent_swarms, max_flights_per_bee, retention_days, auto_archive, default_priority, event_retention_days, trace_retention_days, check_retention_days, webhook_retention_days, auto_maintain, adaptive_enabled, default_token_budget, default_budget_action, cache_enabled, cache_ttl_hours, anomaly_detection_enabled, anomaly_sigma_threshold, anomaly_critical_sigma, anomaly_min_samples, registry_url, registry_cache_hours.Replay a completed, failed, or cancelled swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.--task="..." (override task), --priority=N (override priority), --reset-nectar (start with clean nectar).mcp__hive__hive_swarm_replay with the resolved swarm ID and any overrides./hive-drive on the new swarm.Show fleet-level aggregate metrics.
--period=X from args if present (7d, 30d, 90d, all). Default is "30d".mcp__hive__hive_fleet_metrics with the period.Run data maintenance to clean up old ephemeral data.
--dry-run is given, pass dry_run=true.mcp__hive__hive_maintain with the dry_run flag.Export an installed blueprint as a portable JSON bundle.
mcp__hive__hive_blueprint_export with blueprint_id and optional output_dir.Import a blueprint from a .hive-blueprint.json bundle file.
mcp__hive__hive_blueprint_import with the path.Predict cost and duration for a swarm before starting it.
mcp__hive__hive_swarm_estimate with the blueprint_id.List all pending gated flights with policy details and timeout countdowns.
mcp__hive__hive_gate_list with no params.flight_id | swarm_id | bee_id | policy | gated_at | timeout remaining.Analyze bee performance and recommend parameter adjustments.
--apply is given, set apply=true.mcp__hive__hive_adaptive_tune with the blueprint_id and apply flag.Manually set or override a nectar key on a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_nectar_set with swarm_id, key, and value.Get nectar values from a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_nectar_get with swarm_id and optional key.View blueprint version history.
mcp__hive__hive_blueprint_history with the blueprint_id.version | changes_summary | installed_at.Show structural diff between two versions of a blueprint.
mcp__hive__hive_blueprint_diff with blueprint_id and optional version numbers.View or set token budget for a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_budget_set with the resolved swarm ID and token_budget.
--action=X flag (warn, pause, cancel). Default is "warn".mcp__hive__hive_budget_status with the resolved swarm ID.View cache statistics or clear the cache.
mcp__hive__hive_cache_status.clear, call mcp__hive__hive_cache_clear.
--blueprint=X and --flight=X flags to scope the clear.Compare two swarm runs side-by-side.
mcp__hive__hive_swarm_compare with both identifiers.Inject a flight into a running pipeline.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.--expects=X flag.mcp__hive__hive_flight_inject with swarm_id, after_flight_id, bee_id, input, and optional expects.Skip a pending or waiting flight.
mcp__hive__hive_flight_skip with flight_id and optional reason.Save a swarm configuration as a named template.
--description="...", --priority=N, --variables=key=val,key2=val2.mcp__hive__hive_template_save with name, blueprint_id, and optional description, priority, variables.List all saved swarm templates.
mcp__hive__hive_template_list.name | blueprint_id | description | priority | usage_count | created_at.Start a swarm from a saved template.
--priority=N, --variables=key=val,key2=val2.mcp__hive__hive_template_run with template_name, task, and optional priority/variable overrides.mcp__hive__hive_pollinate with the returned swarm_id to kick off work./hive-drive to run this swarm autonomously."Show DAG visualization for a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_swarm_dag with the resolved swarm ID.Check sub-swarm status for a flight.
mcp__hive__hive_subswarm_status with the flight_id.View model routing decisions.
mcp__hive__hive_routing_history with flight_id.flight_id | bee_id | tier | model | reason | timestamp.List anomaly alerts.
--severity=X (warning, critical) and --status=X (open, acknowledged) from args.mcp__hive__hive_anomaly_alerts with optional filters.alert_id | blueprint | flight | metric | sigma | severity | status.Acknowledge an anomaly alert.
mcp__hive__hive_anomaly_acknowledge with the alert_id.View computed performance baselines.
mcp__hive__hive_anomaly_baselines with it.blueprint | flight | metric | mean | stddev | samples.Check SSE event stream status.
mcp__hive__hive_stream_status.List nectar shares for a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_nectar_shares with the resolved swarm ID.source_swarm | keys | resolved_at.Manually resolve a nectar reference.
latest:blueprint-id, or {{var}}).mcp__hive__hive_nectar_resolve with swarm_id, source, and keys array.Search the blueprint registry.
--url=X for a specific registry URL.mcp__hive__hive_registry_search with query and optional registry_url.blueprint_id | name | description | version | author.Install a blueprint from the registry.
--url=X for a specific registry URL.mcp__hive__hive_registry_install with blueprint_id and optional registry_url.Rate an installed blueprint.
mcp__hive__hive_blueprint_rate with blueprint_id, rating, and optional comment.Create a notification channel.
mcp__hive__hive_channel_create with name, type, and config object.List notification channels.
mcp__hive__hive_channel_list.id | name | type | enabled | created_at.Delete a notification channel.
mcp__hive__hive_channel_delete with channel_id.Create a notification route.
swarm.*, flight.failed).--priority=N.mcp__hive__hive_route_create with channel_id, event_pattern, and optional priority.List notification routes.
mcp__hive__hive_route_list.id | channel | event_pattern | priority.Delete a notification route.
mcp__hive__hive_route_delete with route_id.Create an inbound webhook token.
--expires=ISO8601 for expiry date.mcp__hive__hive_webhook_token_create with name, permissions array, and optional expires_at.List webhook tokens.
mcp__hive__hive_webhook_token_list.id | name | permissions | created_at | expires_at | revoked.Revoke a webhook token.
mcp__hive__hive_webhook_token_revoke with token_id.View inbound webhook audit log.
--limit=N (default 20).mcp__hive__hive_webhook_audit with limit.id | token_id | action | result | ip_address | timestamp.Create a recurring cron-based schedule.
<name> <blueprint_id> <cron_expression> <task_template>. Optional: --overlap=skip|queue|cancel_previous, --priority=N, --vars='{"key":"val"}'.mcp__hive__hive_schedule_create with name, blueprint_id, cron, task_template, and optional params.List recurring schedules.
--blueprint=X and --enabled flags.mcp__hive__hive_schedule_list with filters.id | name | blueprint | cron | enabled | next_run_at | run_count.Delete a schedule.
mcp__hive__hive_schedule_delete with schedule_id.Enable or disable a schedule.
--enable or --disable flag (default: toggle).mcp__hive__hive_schedule_toggle with schedule_id and enabled.View run history for a schedule.
--limit=N.mcp__hive__hive_schedule_history with schedule_id and limit.id | swarm_id | triggered_at | status.Manually evaluate all due schedules.
mcp__hive__hive_schedule_evaluate.List circuit breaker states.
--state=open|closed|half_open.mcp__hive__hive_circuit_list with state.bee_id | state | failure_count | success_count | opened_at.Reset a circuit breaker to closed.
mcp__hive__hive_circuit_reset with bee_id.List dead-lettered flights.
--status=pending|replayed|purged.mcp__hive__hive_dlq_list with filters.id | flight_id | swarm_id | bee_id | last_error | status | created_at.Replay a dead-lettered flight.
mcp__hive__hive_dlq_replay with dead_letter_id.Purge dead letters.
--swarm=N, resolve swarm ID and call with swarm_id.mcp__hive__hive_dlq_purge with the appropriate parameter.Add a blueprint test case.
<blueprint_id> <name>. The user must provide mock_inputs (JSON), mock_outputs (JSON), and assertions (JSON) — prompt if missing.mcp__hive__hive_blueprint_test_add with all parameters.List blueprint test cases.
mcp__hive__hive_blueprint_test_list with blueprint_id.id | name | description | assertion_count.Run blueprint tests.
mcp__hive__hive_blueprint_test_run with test_case_id. Otherwise call with blueprint_id to run the full suite.Delete a blueprint test case.
mcp__hive__hive_blueprint_test_delete with test_case_id.Show the hive health score.
mcp__hive__hive_health.factor | score | weight | detail.Show health score history.
--limit=N (default 20).mcp__hive__hive_health_history with limit.composite_score | trend | computed_at.Tag a swarm with key=value metadata.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.key=value format. Split on the first = to extract the key and value.mcp__hive__hive_swarm_tag with the resolved swarm ID, key, and value.Remove a tag from a swarm.
mcp__hive__hive_swarm_status with it first to resolve the full swarm ID.mcp__hive__hive_swarm_untag with the resolved swarm ID and key.Search swarms with rich filters.
--tag=key:value (filter by tag), --from=ISO8601 (created after), --to=ISO8601 (created before), --status=X, --blueprint=X.--tag=key:value, split on the first : to extract tag_key and tag_value.mcp__hive__hive_swarm_search with query and any parsed filters (tag_key, tag_value, from, to, status, blueprint_id).Create a configuration profile.
key=value arguments.mcp__hive__hive_profile_create with the name and config entries.List all configuration profiles.
mcp__hive__hive_profile_list.name | active | keys | created_at.Activate a configuration profile.
mcp__hive__hive_profile_activate with the name.Delete a configuration profile.
mcp__hive__hive_profile_delete with the name.View bee memories.
feature-dev_worker). If the argument is a swarm number, resolve it first via mcp__hive__hive_swarm_status to identify the swarm, then ask which bee to query.mcp__hive__hive_bee_memory_recall with the bee_id.Clear all memories for a bee.
mcp__hive__hive_swarm_status to identify the swarm, then ask which bee to clear.mcp__hive__hive_bee_memory_forget with the bee_id.Show blueprint dependency graph.
mcp__hive__hive_blueprint_deps with blueprint_id.mcp__hive__hive_blueprint_deps with no params to show dependencies for all installed blueprints.Create an operational playbook (interactive).
swarm.failed, flight.failed, health.degraded, anomaly.critical).notify, restart, escalate, cancel), accepting multiple actions as a list.mcp__hive__hive_playbook_create with name, trigger, threshold, and actions.List operational playbooks.
mcp__hive__hive_playbook_list.id | name | trigger | enabled | last_run | run_count.Delete an operational playbook.
mcp__hive__hive_playbook_delete with the id.Enable or disable an operational playbook.
mcp__hive__hive_playbook_toggle with the id.View playbook execution history.
mcp__hive__hive_playbook_history with id.mcp__hive__hive_playbook_history with no params to show all playbook history.id | playbook_name | trigger_event | actions_taken | result | executed_at.#N for readability.3/7 cells done).stop and resume, always resolve swarm numbers to IDs before calling the action tool.