crewlet

KO | EN

gitlite — search

#react #react-native #machine-learning #developer-tools #mcp #graphql #rag #self-hosted #privacy #awesome-list

(★ 9)

Bring AI agents to life

File Explorer

ZIP Download

- settings.json
- - bug_report.md
  - feature_request.md
- - ci.yml
  - dependabot-merge.yml
  - docs-publish.yml
  - release.yml
- dependabot.yml
- pull_request_template.md
- release.yml
- - backup.go
  - backup_test.go
  - capacity.go
  - capacity_test.go
  - config_cmd.go
  - config_cmd_test.go
  - config_rekey.go
  - config_rekey_test.go
  - configclient.go
  - confluence.go
  - confluence_test.go
  - github.go
  - jira.go
  - listener_test.go
  - llm.go
  - llm_test.go
  - logfile_test.go
  - logsettings_test.go
  - logsink_test.go
  - main.go
  - main_test.go
  - nodeapi.go
  - nodeclient.go
  - ops.go
  - ops_test.go
  - positional.go
  - positional_test.go
  - provision.go
  - provision_test.go
  - reconcile.go
  - reconcile_test.go
  - resolver_test.go
  - retention.go
  - retention_test.go
  - schema_test.go
  - search.go
  - secrets.go
  - secrets_test.go
  - secretsclient.go
  - secretsclient_test.go
  - skillscontainer_test.go
  - slack.go
  - tier_test.go
  - work.go
  - work_test.go
- - favicon.ico
- - main.tsx
- .prettierignore
- .prettierrc.json
- index.html
- package-lock.json
- package.json
- tsconfig.json
- vite.config.ts
- vite.protocol.config.ts
- vitest.config.ts
- - code-sandbox.svg
  - company-as-code.svg
  - crewlet-icon.svg
  - crewlet-mascot.svg
  - hierarchy.svg
  - human-in-loop.svg
  - knowledge.svg
  - README.md
  - turn-engine.svg
- - agent-learning.md
  - agent-runtime.md
  - architecture.md
  - code-sandbox.md
  - configuration.md
  - control-plane.md
  - conversation-sessions.md
  - coordination.md
  - decision-framework.md
  - event-system.md
  - humans-in-the-org.md
  - integration-reconcile.md
  - knowledge-system.md
  - one-on-ones.md
  - organization-model.md
  - overview.md
  - scaling.md
  - scheduling.md
  - seat-ownership.md
  - secret-store.md
  - subscription-llm-backends.md
  - task-engine.md
  - tool-capabilities.md
  - tool-skills.md
  - turn-engine.md
- - ai-authoring.md
  - choosing-your-stack.md
  - configuration.md
  - installation.md
  - quickstart.md
- - backup.md
  - configure-via-api.md
  - consistency.md
  - deployment.md
  - fleet.md
  - replication.md
  - retention.md
  - satellite-nodes.md
  - search.md
  - tools-and-mcp.md
  - work-tracker.md
- - confluence.md
  - datadog.md
  - github.md
  - gitlab.md
  - jira.md
  - mattermost.md
  - slack.md
- - alarms.md
  - api-endpoints.md
  - cli.md
  - dashboard-design.md
  - design-decisions.md
  - environment-variables.md
  - metrics.md
- index.md
- - channel-discovery.md
  - code-runtime.md
  - getting-unstuck.md
  - github.md
  - gitlab.md
  - observed-directives.md
  - platform-mentions.md
  - refine-skill.md
  - reflect-and-persist.md
  - retrieval-research.md
- nimbus-claude-cli.company.yaml
- nimbus-claude-cli.config.yaml
- nimbus.company.yaml
- nimbus.config.yaml
- - channel.go
  - coordstore.go
  - coordstore_test.go
  - service.go
  - service_test.go
  - store_test.go
- - app.go
  - app_test.go
  - backup.go
  - backup_test.go
  - budgets.go
  - budgets_test.go
  - dashboard.go
  - dashboard_internal_test.go
  - dashboard_test.go
  - dashboardjs_test.go
  - health.go
  - mcp.go
  - mcp_test.go
  - otlp.go
  - otlp_test.go
  - purge_test.go
  - query_test.go
  - rest.go
  - rest_test.go
  - retention.go
  - retention_test.go
  - roster.go
  - roster_test.go
  - runtime.go
  - socket_auth_test.go
- - client.go
  - integrationconform_test.go
  - provision.go
  - provision_test.go
  - reconcile.go
  - reconcile_test.go
  - seatcred.go
  - seatcred_test.go
  - setup.go
  - setup_test.go
  - teardown.go
  - teardown_test.go
- - age.go
  - backup.go
  - backup_test.go
  - big_test.go
  - hold_test.go
  - streams.go
- - changefeed.go
  - changefeed_test.go
  - estate_test.go
- - agentmode_test.go
  - bootstrap.go
  - bootstrap_test.go
  - cipher_test.go
  - company.go
  - company_test.go
  - config.go
  - containers.go
  - containers_test.go
  - datadog_test.go
  - embedwidth_test.go
  - envrefs.go
  - envrefs_test.go
  - examples_test.go
  - githubbases_test.go
  - helpers_test.go
  - integrations.go
  - integrations_test.go
  - keys.go
  - learning.go
  - load.go
  - load_test.go
  - logging_test.go
  - mcp.go
  - organization.go
  - provider_keys_test.go
  - providers.go
  - providers_order_test.go
  - redact.go
  - redact_test.go
  - resolve.go
  - resolve_test.go
  - retention.go
  - retention_test.go
  - retiredtracker_test.go
  - rolegithub_test.go
  - roles.go
  - roles_test.go
  - roundtrip_test.go
  - sandbox.go
  - sandbox_pause_test.go
  - sandbox_remote_test.go
  - scheduling.go
  - schema.go
  - schema_test.go
  - skills_container_test.go
  - storedform_test.go
  - storepath_test.go
  - streamcluster_test.go
  - streamsync_test.go
  - turnengine.go
  - unitid_test.go
  - values.go
  - warnings.go
  - warnings_test.go
  - workers.go
  - workers_test.go
- - posture.go
  - posture_test.go
- - client.go
  - cloud_payload_test.go
  - codec.go
  - confluence_test.go
  - hooks.go
  - hooks_test.go
  - integrationconform_test.go
  - mintinto_internal_test.go
  - parse.go
  - payload.go
  - promote.go
  - promote_test.go
  - prompt.go
  - provision.go
  - publish.go
  - publish_test.go
  - search.go
  - setup.go
  - teardown.go
- - backuppoints.go
  - coord.go
  - fleet.go
  - floors.go
  - holds.go
  - keys.go
  - maintenance.go
  - nodestatus.go
  - nodestatus_test.go
  - positions.go
- - client.go
  - client_test.go
  - datadog.go
  - datadog_test.go
  - hooks.go
  - hooks_test.go
  - integrationconform_test.go
  - orphans_test.go
  - parse.go
  - payload.go
  - payload_test.go
  - prompt.go
  - provision.go
  - reconcile.go
  - reconcile_test.go
  - setup.go
  - teardown.go
  - tier.go
- - cluster_test.go
  - company_test.go
  - doc.go
  - gitlab_test.go
  - golden_test.go
  - inbound_test.go
  - learning_test.go
  - logsink_test.go
  - native_test.go
  - prefetch_test.go
  - retention_test.go
  - sandbox_test.go
  - skills_test.go
- - a2a.go
  - a2a_internal_test.go
  - agentrun.go
  - agentrun_internal_test.go
  - backends.go
  - backends_test.go
  - bootcleanup_internal_test.go
  - bridge.go
  - budget.go
  - budget_internal_test.go
  - budgetreport.go
  - budgetreport_internal_test.go
  - capacity.go
  - capacity_internal_test.go
  - coalesce.go
  - confluence.go
  - cooldowns.go
  - cooldowns_internal_test.go
  - cooldowns_test.go
  - datadog.go
  - disconnect.go
  - disconnect_test.go
  - domainfeed.go
  - duty.go
  - duty_internal_test.go
  - embeddings.go
  - embeddings_internal_test.go
  - embedduty.go
  - engine.go
  - engine_test.go
  - epoch.go
  - epoch_internal_test.go
  - equip.go
  - equip_internal_test.go
  - feedsource_test.go
  - github.go
  - githubidentity_internal_test.go
  - githubteardown_internal_test.go
  - gitlab.go
  - gitlabmode_internal_test.go
  - integrationconform_internal_test.go
  - integrations.go
  - integrations_internal_test.go
  - jira.go
  - judge.go
  - learning.go
  - learningbudget.go
  - learningbudget_internal_test.go
  - maintenance.go
  - maintenance_mode.go
  - maintenance_test.go
  - mcp.go
  - mcp_test.go
  - mcpserver_test.go
  - memory.go
  - native.go
  - nativecleanup_internal_test.go
  - nodeidentity_test.go
  - notifications.go
  - notifications_internal_test.go
  - notifications_test.go
  - pagewatchers.go
  - prefetch.go
  - prefetch_internal_test.go
  - promote.go
  - promote_internal_test.go
  - providers.go
  - providers_internal_test.go
  - reanchor.go
  - reconcile.go
  - reconcile_test.go
  - refine_internal_test.go
  - registryrace_internal_test.go
  - reply_internal_test.go
  - republish.go
  - republish_internal_test.go
  - resumesession_internal_test.go
  - retention.go
  - retention_internal_test.go
  - retentioncapacity.go
  - retentioncapacity_test.go
  - retentionreport.go
  - routing.go
  - routing_internal_test.go
  - run.go
  - run_internal_test.go
  - run_test.go
  - runners_internal_test.go
  - sandbox.go
  - sandbox_internal_test.go
  - sandboxcreds_internal_test.go
  - scheduler.go
  - scheduler_test.go
  - sealedsecrets_internal_test.go
  - secrets.go
  - secrets_internal_test.go
  - setuppass.go
  - skills.go
  - slack.go
  - slackteardown.go
  - slackteardown_internal_test.go
  - statelog.go
  - statelogboot_test.go
  - statelogprogress.go
  - statelogprogress_internal_test.go
  - statelogsnapshot_internal_test.go
  - telemetry.go
  - trigger_test.go
  - turn.go
  - turn_test.go
  - turnfailure_internal_test.go
  - watchdog_internal_test.go
  - worksearch.go
  - worksearch_internal_test.go
- - envfile.go
  - envfile_test.go
  - guard_test.go
- - envref.go
  - envref_test.go
- - category.go
  - categorydoc_test.go
  - diagnostic_test.go
  - event.go
  - lossless_test.go
  - payload_test.go
- - fleetsecrets.go
  - fleetsecrets_test.go
  - migrate.go
- - app.go
  - app_test.go
  - client.go
  - findings.go
  - github_test.go
  - hooks.go
  - identity.go
  - identity_test.go
  - integrationconform_test.go
  - mentions.go
  - noregistrar_test.go
  - parse.go
  - payload.go
  - prompt.go
  - provision.go
  - reconcile.go
  - seatapp.go
  - seatapp_test.go
  - seatappread_internal_test.go
  - seatlookup.go
  - seatlookup_test.go
  - setup.go
  - setup_test.go
  - teardown.go
  - teardown_nocred_test.go
  - tier.go
  - tier_test.go
  - transport_internal_test.go
- - admin.go
  - client.go
  - client_test.go
  - findings.go
  - integrationconform_test.go
  - mentions.go
  - mentions_test.go
  - parse.go
  - parse_test.go
  - payload.go
  - prompt.go
  - prompt_test.go
  - provision.go
  - provision_test.go
  - reconcile.go
  - reconcile_test.go
  - setup.go
  - teardown.go
  - teardown_test.go
- - hostbox.go
  - hostbox_test.go
- - httpx.go
  - httpx_test.go
  - limits.go
  - refusal.go
  - refusal_test.go
- - coordstore.go
  - coordstore_test.go
  - finding.go
  - finding_test.go
  - ingress_test.go
  - integration.go
  - integration_test.go
  - report.go
  - report_test.go
  - schedule.go
  - schedule_test.go
  - state.go
  - state_test.go
  - vendorshape_test.go
  - worker.go
  - worker_test.go
- - client.go
  - client_test.go
  - deployment_setup_test.go
  - findings.go
  - hooks.go
  - integrationconform_test.go
  - leads.go
  - leads_test.go
  - noagents_test.go
  - parse.go
  - parse_test.go
  - payload.go
  - payload_test.go
  - prompt.go
  - prompt_test.go
  - provision.go
  - reconcile.go
  - reconcile_test.go
  - setup.go
  - teardown.go
  - teardown_test.go
  - webhooksecret_internal_test.go
- - knowledge.go
  - knowledge_test.go
  - markdown.go
- - background_cluster_test.go
  - cluster.go
  - cluster_test.go
  - counterparty.go
  - counterparty_test.go
  - curator.go
  - curator_test.go
  - diary.go
  - diary_test.go
  - episode.go
  - episode_test.go
  - episodist.go
  - episodist_test.go
  - lifecycle.go
  - lifecycle_test.go
  - modeljson.go
  - modeljson_test.go
  - onboarding.go
  - onboarding_test.go
  - persist.go
  - persist_test.go
  - profiler.go
  - profiler_test.go
  - promote.go
  - promote_test.go
  - rank_internal_test.go
  - recall.go
  - refine.go
  - refine_test.go
  - reflect.go
  - reflect_test.go
  - skill.go
  - skill_test.go
  - skilluse.go
  - skilluse_test.go
  - synthesize.go
  - synthesize_test.go
- - bench_test.go
  - console.go
  - console_test.go
  - file.go
  - file_test.go
  - logging.go
  - logging_test.go
  - trace.go
  - trace_test.go
- - jobs.go
  - maintenance.go
  - maintenance_test.go
  - stubs_test.go
- - api.go
  - client.go
  - client_test.go
  - doctor.go
  - doctor_test.go
  - findings.go
  - findings_test.go
  - fleet.go
  - fleet_test.go
  - integrationconform_test.go
  - parse.go
  - parse_test.go
  - prompt.go
  - prompt_test.go
  - provision.go
  - provision_test.go
  - reconcile.go
  - setup.go
  - setup_test.go
  - teardown.go
  - teardown_test.go
  - transport.go
  - transport_test.go
  - url.go
  - url_test.go
- - annotations.go
  - annotations_test.go
  - bridge.go
  - bridge_test.go
  - catalogue.go
  - catalogue_test.go
  - client.go
  - client_test.go
  - content.go
  - content_test.go
  - doc.go
  - grammar_guard_test.go
  - helper_process_test.go
  - http.go
  - http_test.go
  - instance.go
  - instance_test.go
  - origin.go
  - origin_test.go
  - probe.go
  - probe_test.go
  - spec.go
  - spec_test.go
  - stderr.go
  - stdio.go
  - stdio_test.go
  - testsupport_test.go
  - timeouts.go
  - tool.go
  - tool_test.go
- - concurrency_test.go
  - fleet_test.go
  - gate.go
  - gate_internal_test.go
  - node.go
- - chatprompt.go
  - chatprompt_test.go
  - coalesce.go
  - coalesce_test.go
  - conversation.go
  - conversation_test.go
  - generic.go
  - grammar.go
  - grammar_test.go
  - identity.go
  - identity_test.go
  - log.go
  - phrases.go
  - prompt.go
  - registry.go
  - registry_test.go
  - selfaction.go
  - selfaction_test.go
  - service.go
  - service_test.go
  - status.go
  - status_test.go
  - statuses.go
  - statuses_test.go
  - threads.go
  - threads_test.go
- - categories.go
  - categories_test.go
  - observe_test.go
  - projector.go
  - record.go
  - record_test.go
  - writer.go
- - errors.go
  - hierarchy.go
  - hierarchy_test.go
  - identity.go
  - identity_test.go
  - leads.go
  - organization.go
  - organization_test.go
  - role.go
  - role_test.go
  - schedule.go
  - schedule_test.go
  - unit.go
  - values.go
  - values_test.go
- - apply.go
  - apply_page.go
  - apply_test.go
  - comment.go
  - document.go
  - domain.go
  - domain_test.go
  - feed.go
  - feed_test.go
  - listing_test.go
  - mutation.go
  - pages.go
  - parse.go
  - prompt.go
  - read.go
  - read_test.go
  - record.go
  - roundtrip_test.go
  - rows.go
  - scope.go
  - search.go
  - statelog_test.go
  - store.go
  - store_internal_test.go
  - store_test.go
  - subject.go
  - tables.go
  - write.go
- - procgroup.go
  - procgroup_other.go
  - procgroup_test.go
  - procgroup_unix.go
- - identity.go
  - identity_test.go
  - removal.go
  - secret.go
  - secret_test.go
  - shape_test.go
  - sink.go
  - sinks.go
  - sinks_test.go
- - inflight.go
  - panicline_test.go
  - queue.go
- - allocs_test.go
  - redact.go
  - redact_test.go
- - runtoken.go
- - coordinator.go
  - coordinator_test.go
  - coordstore.go
  - e2b.go
  - e2b_api.go
  - e2b_envd.go
  - e2b_envd_internal_test.go
  - e2b_test.go
  - fake.go
  - idlereader_test.go
  - launch.go
  - launch_test.go
  - local.go
  - local_box.go
  - local_mount.go
  - local_mount_test.go
  - local_test.go
  - manager.go
  - manager_test.go
  - mcprender.go
  - mcprender_test.go
  - otel.go
  - otel_test.go
  - pausettl_test.go
  - pending.go
  - pending_test.go
  - placement.go
  - placement_test.go
  - process_other.go
  - process_unix.go
  - protocol.go
  - setup.go
  - waiter.go
  - waiter_test.go
- - clock_guard_test.go
  - cron.go
  - cron_bench_test.go
  - cron_test.go
  - duty.go
  - duty_test.go
  - entries.go
  - entries_test.go
  - ledger.go
  - memory.go
  - memory_test.go
  - scheduler.go
  - scheduler_test.go
  - sharedclaim.go
  - sharedclaim_test.go
- - apply.go
  - bench_test.go
  - coverage_test.go
  - domain.go
  - embed.go
  - embed_test.go
  - eval.go
  - eval_test.go
  - fanbench_test.go
  - fanout.go
  - fanout_test.go
  - fanwire.go
  - fanwire_test.go
  - fixture.go
  - fuse.go
  - lexical.go
  - lexical_test.go
  - lexicalsearch.go
  - lexicalsources.go
  - localscan.go
  - pagecorpus.go
  - pagecorpus_test.go
  - plan_test.go
  - quantize.go
  - record.go
  - search_test.go
  - semantic.go
  - semantic_test.go
  - shard.go
  - shard_test.go
  - statelog_test.go
  - twostage.go
- - admission_test.go
  - heartbeat.go
  - helpers_test.go
  - host.go
  - host_test.go
  - release_test.go
  - seat.go
  - sweep.go
  - sweep_test.go
  - watchdog.go
  - watchdog_test.go
- - cipher.go
  - document.go
  - document_test.go
  - record.go
  - secrets_test.go
  - sharedtoken.go
  - sharedtoken_test.go
- - marshal_test.go
  - pass.go
  - pass_test.go
  - setup.go
  - setup_test.go
  - shape.go
  - write.go
- - api.go
  - client.go
  - code.go
  - ledger.go
  - manifest.go
  - parse.go
  - parse_test.go
  - prompt.go
  - prompt_test.go
  - provision.go
  - provision_test.go
  - setup.go
  - setup_test.go
  - transport.go
  - transport_internal_test.go
  - transport_test.go
- - adopt.go
  - adopt_test.go
  - adoptrow.go
  - alarmref.go
  - alarmref_test.go
  - alarms.go
  - alarms_test.go
  - alarmtracker.go
  - appender.go
  - apply.go
  - apply_test.go
  - barrier.go
  - barrier_test.go
  - barriercluster_test.go
  - budgets.go
  - budgets_test.go
  - cursorfile.go
  - doc.go
  - domain.go
  - eviction.go
  - eviction_test.go
  - harness_test.go
  - health.go
  - health_test.go
  - identity.go
  - maintenance.go
  - maintenance_test.go
  - outcome.go
  - position.go
  - position_test.go
  - publish.go
  - publish_test.go
  - race_test.go
  - read.go
  - read_test.go
  - readlevels.go
  - readlevels_test.go
  - reanchor.go
  - reanchor_test.go
  - reorder.go
  - report.go
  - report_test.go
  - rows.go
  - scope.go
  - scope_test.go
  - session_test.go
  - snapshot.go
  - snapshot_test.go
  - snapshotrows.go
  - tables.go
  - transfer.go
  - transfer_test.go
  - trim.go
  - trim_test.go
  - vocabulary_test.go
  - waiters.go
- - adopt.go
  - adopt_internal_test.go
  - adopt_test.go
  - agentphases_test.go
  - backup.go
  - backup_test.go
  - boundary.go
  - boundary_test.go
  - capability_test.go
  - caps.go
  - caps_gate_test.go
  - caps_probe_test.go
  - chunk.go
  - chunk_test.go
  - companyconfig.go
  - drainbench_test.go
  - estate_test.go
  - estateboundary_test.go
  - estates_internal_test.go
  - estates_test.go
  - eventlog.go
  - eventwire_test.go
  - faults_test.go
  - libc_linux.go
  - libc_other.go
  - like.go
  - like_test.go
  - lock.go
  - lock_ext_test.go
  - lock_internal_test.go
  - lock_test.go
  - lock_unix.go
  - migrate.go
  - migrate_test.go
  - pinned.go
  - pinned_test.go
  - platform.go
  - quantization_test.go
  - retry_internal_test.go
  - scrub.go
  - scrub_test.go
  - secretvalues.go
  - secretvalues_test.go
  - store.go
  - store_test.go
  - threadfollows.go
  - turnlookup_test.go
  - turso.go
  - turso_test.go
  - tx_test.go
  - writer.go
  - writer_test.go
- - textcut.go
  - textcut_test.go
- - textindex.go
  - textindex_test.go
- - tokens.go
  - tokens_test.go
- - delivers_test.go
  - registry.go
  - surface.go
  - surface_internal_test.go
  - tools_test.go
- - meter.go
  - meter_test.go
  - span.go
  - tracing.go
  - tracing_test.go
- - activityread.go
  - activityread_test.go
  - apply.go
  - apply_closure.go
  - apply_history.go
  - apply_objects.go
  - apply_task.go
  - apply_test.go
  - caps_test.go
  - catalogue.go
  - catalogue_test.go
  - catalogueread.go
  - changekind_test.go
  - chart.go
  - chart_test.go
  - coerce.go
  - coerce_test.go
  - config.go
  - config_test.go
  - dates.go
  - dates_test.go
  - depend.go
  - depend_test.go
  - detail.go
  - detail_test.go
  - detailfields.go
  - documents.go
  - domain.go
  - duty.go
  - duty_test.go
  - evictions.go
  - expand.go
  - expand_test.go
  - fallbackkind_test.go
  - feed.go
  - feed_test.go
  - fieldquery.go
  - fieldquery_test.go
  - fields_test.go
  - fieldvalues.go
  - gates.go
  - gates_test.go
  - goals.go
  - goals_test.go
  - goalsread.go
  - grouping.go
  - grouping_test.go
  - inboxread.go
  - inboxread_test.go
  - inboxsweep.go
  - inboxsweep_test.go
  - maximal_test.go
  - merge_test.go
  - mutation.go
  - mywork.go
  - mywork_test.go
  - onesided.go
  - paging_test.go
  - parity_test.go
  - parse.go
  - parse_test.go
  - person.go
  - person_test.go
  - personread.go
  - plan_test.go
  - policy.go
  - policy_test.go
  - precondition_test.go
  - producers_test.go
  - project.go
  - project_test.go
  - projectsread.go
  - projectsread_test.go
  - prompt.go
  - prompt_test.go
  - promptobject.go
  - query.go
  - query_test.go
  - rank.go
  - rank_test.go
  - read.go
  - read_test.go
  - reanchor.go
  - recipients.go
  - recipients_test.go
  - record.go
  - record_test.go
  - relate.go
  - roundtrip_test.go
  - rows.go
  - schema_test.go
  - scope_test.go
  - search.go
  - sequence.go
  - sequence_test.go
  - session_test.go
  - sprintduty_test.go
  - sprints.go
  - sprints_test.go
  - sprintsread.go
  - sprintsread_test.go
  - staleness_test.go
  - statelog_test.go
  - subject.go
  - subject_test.go
  - subtasks_test.go
  - tables.go
  - tags.go
  - tags_test.go
  - thread.go
  - tools.go
  - totals.go
  - totals_test.go
  - trash.go
  - trash_test.go
  - unblock.go
  - unblock_test.go
  - views.go
  - views_test.go
  - viewsread.go
  - wake.go
  - wakes_test.go
  - walk.go
  - walk_test.go
  - watch_test.go
  - write.go
- - version.go
  - version_test.go
- - whsec.go
  - whsec_test.go
- - workkey.go
  - workkey_test.go
- bootstrap.schema.json
- company.schema.json
- README.md
- check-signoff.sh
- check-signoff_test.sh
- gitlab-dev-bootstrap.sh
- mattermost-dev-bootstrap.sh
- - SKILL.md
- - favicon.ico
  - index.html
  - protocol.js
- crewlet-icon.svg
- static.go
.env.example
.gitignore
.golangci.yml
.goreleaser.yaml
CLAUDE.md
CODE_OF_CONDUCT.md
CONTRIBUTING.md
DCO
docker-compose.yml
Dockerfile
go.mod
go.sum
LICENSE
Makefile
README.md
RELEASING.md
SECURITY.md

# Use via CDN

jsDelivr

jsDelivr serves any public GitHub repository as a CDN with zero setup. Pick a version and a file to get a ready-to-paste link and snippet.

Version

File

Link

Example

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

**Run an AI agent company.**

Crewlet is an open-source Go engine for orchestrating hierarchically organized
AI agent companies. You describe a company in YAML — mission, org chart, roles,
policies, integrations — and Crewlet runs it: one persistent agent per seat,
planning and executing real work in chat, your issue tracker, and your code host,
learning from what it did, and escalating to the humans in the org chart when stuck.

[![Release](https://img.shields.io/github/v/release/crewlet/crewlet?style=flat-square&color=7c56ff&label=release)](https://github.com/crewlet/crewlet/releases)
[![CI](https://github.com/crewlet/crewlet/actions/workflows/ci.yml/badge.svg)](https://github.com/crewlet/crewlet/actions/workflows/ci.yml)
[![License: MIT](https://img.shields.io/badge/license-MIT-7c56ff?style=flat-square)](https://raw.githubusercontent.com/crewlet/crewlet/main/LICENSE)
[![Go 1.27+](https://img.shields.io/badge/go-1.27%2B-7c56ff?style=flat-square)](https://raw.githubusercontent.com/crewlet/crewlet/main/go.mod)
[![Docs](https://img.shields.io/badge/docs-read-7c56ff?style=flat-square)](https://raw.githubusercontent.com/crewlet/crewlet/main/docs/index.md)

<a href="docs/getting-started/quickstart.md">Quickstart</a> ·
<a href="docs/getting-started/choosing-your-stack.md">Choosing your stack</a> ·
<a href="docs/concepts/overview.md">Concepts</a> ·
<a href="docs/index.md">Documentation</a> ·
<a href="CONTRIBUTING.md">Contributing</a>

</div>

---

## The org chart is the execution graph

Identity, delegation, communication, and knowledge are all scoped by where a seat
sits in the hierarchy. The chart you draw is the graph the engine runs.

<table>
<tr>
<td width="33%" align="center" valign="top">
<img src="docs/assets/hierarchy.svg" width="64" alt=""> 
The org chart runs the work 
Departments, teams, and seats nest to any depth. Leads assign work by reasoning
about their reports — not by a routing algorithm.
</td>
<td width="33%" align="center" valign="top">
<img src="docs/assets/company-as-code.svg" width="64" alt=""> 
Company as code 
The whole company is versioned YAML the engine stores and serves — live-editable
through a REST API, with no restart for role, provider, or integration changes.
</td>
<td width="33%" align="center" valign="top">
<img src="docs/assets/turn-engine.svg" width="64" alt=""> 
Execute → Review 
One agentic loop decides and acts with everything it has read still in front of
it, then an adversarial reviewer judges the record — and loops it back when the
work isn't done.
</td>
</tr>
<tr>
<td width="33%" align="center" valign="top">
<img src="docs/assets/code-sandbox.svg" width="64" alt=""> 
Agents that ship code 
An engineer seat runs a coding agent in an isolated sandbox and opens the merge
request under its own identity — not yours.
</td>
<td width="33%" align="center" valign="top">
<img src="docs/assets/knowledge.svg" width="64" alt=""> 
Memory that compounds 
Agents search the team knowledge base at query time, keep a private diary, recall
similar past turns, and synthesize reusable skills.
</td>
<td width="33%" align="center" valign="top">
<img src="docs/assets/human-in-loop.svg" width="64" alt=""> 
Humans hold seats too 
Put yourself in the chart. Human seats lead units and receive escalations on the
surfaces you already use.
</td>
</tr>
</table>

---

## How a turn works

A trigger — a chat message, a work-item webhook, a schedule — wakes exactly one
agent, which runs a two-stage turn:

```mermaid
flowchart LR
 T["Trigger chat · work item · schedule"] --> E
 E["Execute one loop: discover tools, act, then say what it did"] --> R["Review judge the record: what actually ran"]
 R -->|done| S["Work shipped: comments, MRs, docs"]
 R -->|self_iterate| E
```

Deciding and acting are ONE conversation, so a tool result the agent read is still
in front of it when it acts on that result — and it discovers the tools it needs
as it goes instead of naming them in advance. The reviewer is a separate, narrower
prompt on its own model — a frontier model to work, a cheap one to check. See
[Turn Engine](docs/concepts/turn-engine.md).

---

## Quickstart

There is no infrastructure to bring up. Crewlet is one binary: it embeds its
event stream (a NATS JetStream server) and its database is a local file it
creates. Describe a company and run it:

```bash
go install github.com/crewlet/crewlet/cmd/crewlet@latest
# or grab a signed binary from the releases page, or run
# ghcr.io/crewlet/crewlet

export CREWLET_API_TOKEN_FOUNDER="$(openssl rand -hex 32)"
export ANTHROPIC_API_KEY="sk-ant-..."     # or OpenAI, or any OpenAI-compatible endpoint
export OPENAI_API_KEY="sk-..."            # embeddings

./crewlet run          # reads ./crewlet.yaml (Tier A) and ./company.yaml (Tier B)
```

Those two filenames are the defaults, so nothing needs naming when they sit in
the working directory; `-config` and `-company` point elsewhere. The Tier A
file is the short one — where this node's store, stream and API live — and the
[quickstart](docs/getting-started/quickstart.md) writes it out in full.

The company file is a **bootstrap seed**: it fills an empty store and is
ignored once a company exists, so a restart never reverts a change made live.
The store is the source of truth at runtime. To change a running fleet, run
`crewlet config import company.yaml` — it goes through the node's API and every
node converges with no restart. To make the file win on a restart instead, pass
`-import-company`.

A four-agent company is about 40 lines of YAML:

```yaml
name: "Acme AI"
mission: "Ship AI-powered products fast"

providers:
  llm:
    default:
      type: anthropic
      model: claude-sonnet-5
      api_keys: ["${ANTHROPIC_API_KEY}"]

roles:
  - name: CEO
    goal: "Set product vision and make final calls"
    manages: [CTO, PM]

units:
  - name: Core Engineering
    type: team
    lead: CTO
    roles:
      - name: CTO
        goal: "Set technical direction and unblock engineers"
        manages: [Engineer]
      - name: Engineer
        goal: "Implement features, write tests, ship quality code"
```

The dashboard and webhook API come up with the engine. The full
[Quickstart](docs/getting-started/quickstart.md) walks through watching an agent's
first turn with no integrations at all, then wiring in the real ones.

> **Want a whole company to run?** Two complete seven-seat reference orgs ship
> in `examples/`, the same company at both ends of the stack.
> `nimbus.company.yaml` is the full one — GitLab, Mattermost, a metered key,
> and the engine's own work tracker and knowledge base.
> `nimbus-claude-cli.company.yaml` is the short path: chat on Mattermost, its
> model a coding CLI you already subscribe to, and its engineering seats running
> that same CLI as their own agentic loop in a sandbox on the engine host, on
> that same login — no API key, and no service beyond the chat server in this
> repo's compose file. Neither needs a tracker or a wiki account: both run the
> engine's own, and [Jira](docs/integrations/jira.md) /
> [Confluence](docs/integrations/confluence.md) are the blocks that move those
> halves to Atlassian. Every setting in both carries its reasoning in a comment.

> **Rather not write it by hand?** An AI assistant can interview you and author
> both files, checking its own work against the shipped JSON Schema — see
> [Authoring with an AI assistant](docs/getting-started/ai-authoring.md) for a
> step-by-step walkthrough.

---

## Plug in your stack

Crewlet is the engine; the surfaces your agents work on are yours to choose — see
[Choosing your stack](docs/getting-started/choosing-your-stack.md).

**Routed end to end** — a delivery from one of these wakes the seat it concerns:

| | Options |
|---|---|
| **LLM** | [Anthropic](docs/getting-started/quickstart.md#llm-options), OpenAI, or **any OpenAI-compatible endpoint** — including your own vLLM / LiteLLM gateway |
| **Tracker** | [Jira](docs/integrations/jira.md) — Cloud or Data Center — or [GitLab](docs/integrations/gitlab.md) / [GitHub](docs/integrations/github.md) issues |
| **Knowledge base** | [Confluence](docs/integrations/confluence.md) — searched at turn start and on demand, run as the asking seat |
| **Code host** | [GitLab](docs/integrations/gitlab.md) — gitlab.com or self-hosted — or [GitHub](docs/integrations/github.md), github.com or Enterprise Server |
| **Chat** | [Mattermost](docs/integrations/mattermost.md) — self-hosted, one bot identity per agent — or [Slack](docs/integrations/slack.md), one app per agent |
| **Code sandbox** | [The engine host](docs/concepts/code-sandbox.md), as a process tree or a container; Claude Code or OpenCode as the coding agent |

The knowledge base is **single-homed** — one searcher per company, because two would
make an agent's answer to "what do we already know about this" depend on which one was
asked. It sits behind the backend-neutral `knowledge.Searcher` seam, so a second
backend is a new implementation rather than a rewrite of everything that searches.
Everything else composes: a company can run GitHub beside GitLab, or Slack beside
Mattermost, which is what a migration and an open-source presence both look like.

One command provisions the whole fleet — a bot or
service account per seat, memberships, per-agent tokens minted into your config's own
`${VAR}` references, and the webhooks:

```bash
crewlet mattermost provision company.yaml
crewlet gitlab provision company.yaml -public-url <url>
crewlet jira provision company.yaml -public-url <url> -env-file .env
crewlet slack provision company.yaml -public-url <url> -env-file .env
crewlet github provision company.yaml -public-url <url> -env-file .env
```

What each command can actually do differs by what the vendor allows: Mattermost and
GitLab create an account per seat and mint its token; Jira and GitHub issue no
credential on a provisioner's behalf, so those runs report which account each seat's
own credential authenticates as and register the webhooks.

Mattermost takes no URL because nothing has to reach the engine: it holds one
outbound websocket per agent seat instead of receiving webhooks, so that loop
runs behind NAT with no tunnel.

Add `-secret-store` and the minted credentials go straight into the encrypted
[secret store](docs/concepts/secret-store.md) the engine reads `${VAR}` from —
no env file to source, no shell to be in.

---

## Architecture

```mermaid
flowchart LR
 EXT["External surfaces Mattermost / Slack · Jira / Confluence GitHub / GitLab"]
 API["REST API + dashboard embedded, or its own process"]
 Q["Event stream embedded JetStream, clustered or external for a fleet"]
 ENG["Engine one turn engine per seat"]
 DB[("Store one local file events · memory · config")]

EXT -->|webhooks| API
 API <--> Q
 Q <--> ENG
 ENG -->|"MCP tools, as each agent"| EXT
 ENG --> DB
 API --> DB
```

That is the glance; **[Architecture](docs/concepts/architecture.md)** draws the
same engine at six zoom levels — the process boundary, one node's interior, the
path a trigger takes across a fleet, the path a turn takes, where every table
and bucket lives, and what changes when a second node appears.

Agents are callback-driven: messages on an agent's inbox topic invoke its handler,
and events that piled up while it was busy are batched into a single digest turn.
Config lives in two tiers — an ops-owned bootstrap file on disk, and a versioned,
live-editable company document in the store that can be
[encrypted at rest](docs/concepts/configuration.md#secrets) as a single opaque blob.

---

## CLI

```bash
crewlet run                                   # boot; seeds the store from company.yaml
crewlet run -roles ingress                    # API + webhooks only (split deployments)
crewlet validate                              # check both tiers before booting
crewlet schema company                        # JSON Schema (editors, CI, agents)

crewlet config import | show | export | revisions | diff | activate
crewlet secrets keygen                        # a keyring key — installing it seals the config
crewlet secrets set LLM_API_KEY               # a secret the engine reads ${VAR} from

crewlet confluence import company.yaml docs/  # publish knowledge docs + tool skills
crewlet gitlab provision company.yaml         # reconcile the company's seats into GitLab
```

Full reference: [CLI](docs/reference/cli.md) ·
[API endpoints](docs/reference/api-endpoints.md) ·
[Environment variables](docs/reference/environment-variables.md).

---

## Documentation

**Start here**
- [Installation](docs/getting-started/installation.md) — prerequisites and infrastructure
- [Quickstart](docs/getting-started/quickstart.md) — a company running in minutes
- [Choosing your stack](docs/getting-started/choosing-your-stack.md) — every external dependency, and what you set up yourself
- [Authoring with an AI assistant](docs/getting-started/ai-authoring.md) — let an AI write your company config
- [Configuration reference](docs/getting-started/configuration.md) — every YAML field

**Core concepts**
- [Overview](docs/concepts/overview.md) · [Organization model](docs/concepts/organization-model.md) · [Humans in the org chart](docs/concepts/humans-in-the-org.md)
- [Agent runtime](docs/concepts/agent-runtime.md) · [Turn engine](docs/concepts/turn-engine.md) · [Code sandbox](docs/concepts/code-sandbox.md)
- [Knowledge system](docs/concepts/knowledge-system.md) · [Agent learning](docs/concepts/agent-learning.md) · [Tool skills](docs/concepts/tool-skills.md)
- [Event system](docs/concepts/event-system.md) · [Scheduling](docs/concepts/scheduling.md) · [Configuration](docs/concepts/configuration.md)

**Guides**
- [Tools & MCP](docs/guides/tools-and-mcp.md) · [Deployment](docs/guides/deployment.md) · [Running a Fleet](docs/guides/fleet.md)

</td>
<td valign="top" align="center" width="38%">
<img src="docs/assets/crewlet-mascot.svg" width="300" alt="">
</td>
</tr>
</table>

---

## Contributing

Issues and pull requests are welcome. [CONTRIBUTING.md](CONTRIBUTING.md) covers the
dev setup, conventions, and the checks CI runs:

```bash
make build    # go build ./...
make check    # every gate CI runs: gofmt, go mod tidy -diff, the DCO
              #   sign-off, vet, golangci-lint, build, the race suite,
              #   and a cross-compile of every release target
make help     # everything else
```

Commits are signed off (`git commit -s`), which certifies the
[DCO](DCO) over them — CI checks it.

Releases are cut by pushing a `v*` tag — see [RELEASING.md](RELEASING.md).

Security issues: please report them privately — see [SECURITY.md](SECURITY.md).

## License

[MIT](LICENSE)

The dashboard's icon sprite is adapted from
[Feather Icons](https://feathericons.com) — MIT License,
Copyright (c) 2013-2023 Cole Bemis.

crewlet

File Explorer

# Use via CDN

Was this content helpful?

Recent Feedback

Download README