npx claudepluginhub codota/ctx-customer-pack-distributable --plugin ctx-customer-pack_builtin tools: get_coding_guidelines, get_cve_resolution_status, query_entities, query_symbols, search_knowledge, search_skills
Agent orchestration tools: get_agent_run_output, get_agent_run_status, invoke_agent, list_agent_kinds
Ai readiness tools: get_ai_readiness_history, get_ai_readiness
Architecture tools: get_runbook, get_security_patterns, search_adrs, search_flows, search_incidents
Assess the blast radius of a change — what services, flows, and teams are affected.
Change confidence tools: get_change_confidence, get_file_risk_factors
Plan and execute code migrations — find equivalent fields, get migration examples.
Collect debug logs, checkpoints, and state into a redacted support bundle (tar.gz). Use when something goes wrong and the user needs to send diagnostics to support.
Composite skills tools: blast_radius, code_migration, dependency_check, incident_response, investigate_service, understand_flow
Confluence tools: add_confluence_comment, create_confluence_page, get_confluence_page, search_confluence_pages, update_confluence_page
Query the Context Engine knowledge graph — investigate services, check blast radius, search entities, manage issues, and assess change risk.
Decision context tools: create_code_module, get_file_context, list_code_modules, resolve_file_to_service
Check dependency health, find vulnerabilities, plan migrations.
Development tools: create_feature_pr, get_feature, merge_feature_pr, start_new_feature, update_feature_decisions
Documentation tools: get_service_documentation, list_documentation, search_all_documentation
Flows tools: get_flow_services, get_flow, get_service_flows, list_flows
Get workflow guides for using Context Engine tools effectively. Call this at the start of a session to understand which tools to use for common workflows like investigating services, responding to incidents, assessing blast radius, managing dependencies, and exploring architecture. Returns a comprehensive guide covering all major workflows with step-by-step instructions.
Git insights tools: get_author_expertise, get_codebase_hotspots, get_coupling_issues, get_file_experts, get_file_risk, get_git_insights_summary, get_module_boundaries, get_recent_activity, get_related_files
Gitlab tools: add_gitlab_mr_comment, create_gitlab_merge_request, search_gitlab_issues
Graph tools tools: find_entities, get_entity_by_id, traverse_edges
Get incident response context — runbooks, escalation contacts, recent incidents for a service.
Deep-dive investigation of a service — dependencies, dependents, ownership, ADRs, incidents.
Jfrog tools: check_dependency_health, get_adoption_status, get_cve_blast_radius, get_migration_examples, get_migration_path, get_migration_status, get_package_usage_patterns, get_package_vulnerabilities, get_recommended_packages, get_transitive_dependents, list_all_packages, list_all_vulnerabilities, search_internal_packages, test_cypher_tenant, test_cypher_type, test_cypher
Manage Jira issues — get, create, transition, and comment on issues.
Linear tools: add_linear_comment, create_linear_issue, list_linear_issues, list_linear_teams, transition_linear_issue, update_linear_issue
Lspmcp tools: semantic_find_referencing_symbols, semantic_find_symbol, semantic_get_symbols_overview, semantic_insert_after_symbol, semantic_list_dir, semantic_read_file, semantic_replace_symbol_body, semantic_search_for_pattern
Walk through the complete 7-step Context Engine onboarding methodology. Validates setup, builds a testing lab, loads project data, measures baseline performance, enriches with domain context, and generates a rollout plan.
Opsgenie tools: acknowledge_opsgenie_alert, add_opsgenie_note, close_opsgenie_alert, escalate_opsgenie_alert, get_opsgenie_alert
Ownership tools: get_all_teams, get_code_reviewers, get_incident_contacts, get_service_ownership, get_team_services
Pagerduty tools: acknowledge_pagerduty_incident, add_pagerduty_note, resolve_pagerduty_incident
Query GCP Error Reporting API for error group statistics with time-windowed counts. Returns timedCounts arrays for spike detection. Requires a connected and enabled gcp_monitoring data source.
Review pull requests with architectural context — risk scores, ADRs, hotspots, ownership.
Search the knowledge graph for entities, documentation, and patterns.
Semantic concepts tools: find_equivalent_fields, get_concept_implementations, get_domain_concepts, list_semantic_domains, search_concepts
Servicenow tools: add_servicenow_work_note, create_servicenow_change_request, create_servicenow_incident, list_servicenow_incidents, update_servicenow_incident
Services tools: get_service_context, get_service_dependencies, get_service_dependents, get_service, list_services
Slack tools: post_slack_message, update_slack_message
Understand a business flow end-to-end — services involved, ADRs, incidents.
Upgrades tools: get_major_upgrades, get_package_upgrade_history, get_upgrade_summary, get_upgrades_with_code_changes, get_upgrades_without_code_changes, list_upgrades
Confluence tools: add_confluence_comment, create_confluence_page, get_confluence_page, search_confluence_pages, update_confluence_page
Query the Context Engine knowledge graph — investigate services, check blast radius, search entities, manage issues, and assess change risk.
Decision context tools: create_code_module, get_file_context, list_code_modules, resolve_file_to_service
Check dependency health, find vulnerabilities, plan migrations.
Development tools: create_feature_pr, get_feature, merge_feature_pr, start_new_feature, update_feature_decisions
Documentation tools: get_service_documentation, list_documentation, search_all_documentation
Flows tools: get_flow_services, get_flow, get_service_flows, list_flows
Get workflow guides for using Context Engine tools effectively. Call this at the start of a session to understand which tools to use for common workflows like investigating services, responding to incidents, assessing blast radius, managing dependencies, and exploring architecture. Returns a comprehensive guide covering all major workflows with step-by-step instructions.
Git insights tools: get_author_expertise, get_codebase_hotspots, get_coupling_issues, get_file_experts, get_file_risk, get_git_insights_summary, get_module_boundaries, get_recent_activity, get_related_files
Gitlab tools: add_gitlab_mr_comment, create_gitlab_merge_request, search_gitlab_issues
Graph tools tools: find_entities, get_entity_by_id, traverse_edges
_builtin tools: get_coding_guidelines, get_cve_resolution_status, query_entities, query_symbols, search_knowledge, search_skills
Agent orchestration tools: get_agent_run_output, get_agent_run_status, invoke_agent, list_agent_kinds
Ai readiness tools: get_ai_readiness_history, get_ai_readiness
Architecture tools: get_runbook, get_security_patterns, search_adrs, search_flows, search_incidents
Assess the blast radius of a change — what services, flows, and teams are affected.
Change confidence tools: get_change_confidence, get_file_risk_factors
Plan and execute code migrations — find equivalent fields, get migration examples.
Collect debug logs, checkpoints, and state into a redacted support bundle (tar.gz). Use when something goes wrong and the user needs to send diagnostics to support.
Composite skills tools: blast_radius, code_migration, dependency_check, incident_response, investigate_service, understand_flow
Get incident response context — runbooks, escalation contacts, recent incidents for a service.
Deep-dive investigation of a service — dependencies, dependents, ownership, ADRs, incidents.
Jfrog tools: check_dependency_health, get_adoption_status, get_cve_blast_radius, get_migration_examples, get_migration_path, get_migration_status, get_package_usage_patterns, get_package_vulnerabilities, get_recommended_packages, get_transitive_dependents, list_all_packages, list_all_vulnerabilities, search_internal_packages, test_cypher_tenant, test_cypher_type, test_cypher
Manage Jira issues — get, create, transition, and comment on issues.
Linear tools: add_linear_comment, create_linear_issue, list_linear_issues, list_linear_teams, transition_linear_issue, update_linear_issue
Lspmcp tools: semantic_find_referencing_symbols, semantic_find_symbol, semantic_get_symbols_overview, semantic_insert_after_symbol, semantic_list_dir, semantic_read_file, semantic_replace_symbol_body, semantic_search_for_pattern
Walk through the complete 7-step Context Engine onboarding methodology. Validates setup, builds a testing lab, loads project data, measures baseline performance, enriches with domain context, and generates a rollout plan.
Opsgenie tools: acknowledge_opsgenie_alert, add_opsgenie_note, close_opsgenie_alert, escalate_opsgenie_alert, get_opsgenie_alert
Ownership tools: get_all_teams, get_code_reviewers, get_incident_contacts, get_service_ownership, get_team_services
Pagerduty tools: acknowledge_pagerduty_incident, add_pagerduty_note, resolve_pagerduty_incident
Query GCP Error Reporting API for error group statistics with time-windowed counts. Returns timedCounts arrays for spike detection. Requires a connected and enabled gcp_monitoring data source.
Review pull requests with architectural context — risk scores, ADRs, hotspots, ownership.
Search the knowledge graph for entities, documentation, and patterns.
Semantic concepts tools: find_equivalent_fields, get_concept_implementations, get_domain_concepts, list_semantic_domains, search_concepts
Servicenow tools: add_servicenow_work_note, create_servicenow_change_request, create_servicenow_incident, list_servicenow_incidents, update_servicenow_incident
Services tools: get_service_context, get_service_dependencies, get_service_dependents, get_service, list_services
Slack tools: post_slack_message, update_slack_message
Understand a business flow end-to-end — services involved, ADRs, incidents.
Upgrades tools: get_major_upgrades, get_package_upgrade_history, get_upgrade_summary, get_upgrades_with_code_changes, get_upgrades_without_code_changes, list_upgrades
Context Engine skills, data loader, and onboarding methodology for AI coding agents.
| Package | What it includes | Depends on |
|---|---|---|
| core | 37 skills, hooks, MCP proxy for your AI agent | — |
| loader | tabnine-ctx-loader CLI for bulk data loading | core |
| onboarder | tabnine-ctx-onboard CLI + 7-step evaluation methodology | core, loader |
Core only — skills for your AI agent:
curl -fsSL https://raw.githubusercontent.com/codota/ctx-customer-pack-distributable/main/installers/install.sh | bash -s -- --package core --agent claude
Core + data loader CLI:
curl -fsSL https://raw.githubusercontent.com/codota/ctx-customer-pack-distributable/main/installers/install.sh | bash -s -- --package loader --agent claude
Everything (core + loader + onboarding):
curl -fsSL https://raw.githubusercontent.com/codota/ctx-customer-pack-distributable/main/installers/install.sh | bash -s -- --package all --agent claude
The installer fetches files directly from GitHub — no clone needed. Dependency resolution is automatic: loader includes core, all includes everything. Replace claude with cursor, gemini, or tabnine for other agents.
claude plugin install --from https://github.com/codota/ctx-customer-pack-distributable
Installs the full plugin bundle via the marketplace manifest: 37 skills + hooks (decision-context, change-confidence) + MCP server (tabnine-ctx-cloud). The loader and onboarder are CLI tools — use the curl | bash installer above to add them.
export CTX_API_URL=https://ctx.your-company.com
export CTX_API_KEY=ctx_your_key_here
| Variable | Required | Description |
|---|---|---|
CTX_API_URL | Yes | Context Engine server URL |
CTX_API_KEY | Yes | API key (never pass as CLI flag) |
HTTPS_PROXY | No | HTTP proxy for corporate networks |
NODE_EXTRA_CA_CERTS | No | Custom CA certificate path |
ctx-skills, tabnine-ctx-loader, and tabnine-ctx-onboardtabnine-ctx-loader init --template github-jira-slack --output tabnine-ctx-loader.yaml
tabnine-ctx-loader validate --manifest tabnine-ctx-loader.yaml --json
tabnine-ctx-loader load --manifest tabnine-ctx-loader.yaml --json
tabnine-ctx-loader status --json
Templates in loader/templates/.
tabnine-ctx-onboard step-0 --json # Validate setup
tabnine-ctx-onboard step-1 --repo-path . --json # Generate test plan
tabnine-ctx-onboard step-2 --manifest m.yaml # Load data
tabnine-ctx-onboard step-3 --json # Baseline (no MCP)
tabnine-ctx-onboard step-4 --json # Baseline (with MCP)
tabnine-ctx-onboard step-5 --repo-path . --json # Domain enrichment
tabnine-ctx-onboard step-6 --json # Measure with domain
tabnine-ctx-onboard step-7 --json # Rollout plan
| Agent | Tier | Core install |
|---|---|---|
| Claude Code | 1 | Plugin or file copy |
| Cursor | 1 | File copy (skills + rules) |
| Gemini CLI | 1 | File copy |
| Tabnine | 1 | File copy (global) |
| GitHub Copilot | 2 (preview) | File copy |
| Codex | 2 (preview) | File copy |
core/agents/ 37 skills × 6 agents + hooks + MCP
core/tool-schemas.json
loader/bin/tabnine-ctx-loader
loader/templates/ 3 manifest templates
onboarder/bin/tabnine-ctx-onboard
onboarder/skills/ 7 onboarding step skills
installers/install.sh
Standalone image generation plugin using Nano Banana MCP server. Generates and edits images, icons, diagrams, patterns, and visual assets via Gemini image models. No Gemini CLI dependency required.
Admin access level
Server config contains admin-level keywords
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Persistent memory system for Claude Code - seamlessly preserve context across sessions
Qiushi Skill: methodology skills for AI agents guided by seeking truth from facts, with Claude Code, Cursor, OpenClaw, Codex, OpenCode, and Hermes guidance.
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Comprehensive UI/UX design plugin for mobile (iOS, Android, React Native) and web applications with design systems, accessibility, and modern patterns
Search and retrieve documents from local markdown files.
Modifies files
Hook triggers on file write and edit operations