Tool Reference
ido4 provides MCP tools organized into categories. The exact count varies by methodology profile because container management tools and transition tools are generated dynamically from the profile.
| Profile | Tools | Reason |
|---|---|---|
| Hydro | 58 | 2 containers (Wave, Epic) + 9 transitions |
| Scrum | 56 | 2 containers (Sprint, Epic) + 8 transitions |
| Shape Up | 54 | 3 containers (Cycle, Bet, Scope) + 10 transitions |
Every write tool supports dryRun, goes through BRE validation, and creates an audit entry.
dryRun: true — Validates the operation fully (runs the entire BRE pipeline) and returns what would happen, but doesn’t mutate GitHub or create audit events. Use this to preview impact before committing to a transition.
Tools are governed operations, not API wrappers. Every write goes through: input validation → BRE pipeline (34 steps) → GitHub mutation → audit event → structured response with suggestions. This is why an agent can’t accidentally violate a governance principle.
Task Governance
Workflow Transitions (dynamic)
Transition tools are generated from the active profile’s transition definitions. The exact set depends on which transitions the profile defines.
Universal transitions (all profiles):
| Tool | Description | Key Parameters |
|---|---|---|
start_task | Begin work — validates dependencies, container assignment, singularity | issueNumber, dryRun? |
review_task | Submit for review/QA | issueNumber, dryRun? |
block_task | Block a task with a required reason | issueNumber, reason, dryRun? |
unblock_task | Unblock a task | issueNumber, dryRun? |
return_task | Backward transition to a previous status | issueNumber, targetStatus, reason, dryRun? |
Hydro-specific:
| Tool | Description |
|---|---|
refine_task | Backlog -> In Refinement |
ready_task | In Refinement/Backlog -> Ready for Dev |
approve_task | In Review -> Done |
complete_task | Administrative completion (Done -> Done) |
Scrum-specific:
| Tool | Description |
|---|---|
plan_task | Product Backlog -> Sprint Backlog |
approve_task | In Review -> Done |
Shape Up-specific:
| Tool | Description |
|---|---|
shape_task | Raw Idea -> Shaped |
bet_task | Shaped -> Bet On |
ship_task | QA -> Shipped |
kill_task | Building/QA/Blocked -> Killed (circuit breaker) |
All transition tools accept optional message (added as issue comment), context (structured development context written as ido4 context comment), and skipValidation (bypasses BRE — recorded in audit trail).
Task Data (4 tools)
| Tool | Description | Key Parameters |
|---|---|---|
get_task | Full task details: status, containers, dependencies, all metadata | issueNumber |
get_task_field | Get a specific field value | issueNumber, field? |
list_tasks | List all tasks with optional filtering by status, container, assignee | status?, wave?, assignee? |
create_task | Create a new task (GitHub issue) and add to the project board | title, body?, containers, metadata fields |
Task Intelligence (6 tools)
| Tool | Description | Key Parameters |
|---|---|---|
validate_transition | Dry-run a specific transition | issueNumber, transition |
validate_all_transitions | Check all possible transitions — shows what’s allowed and what’s blocked | issueNumber |
find_task_pr | Find the pull request linked to a task | issueNumber |
get_pr_reviews | Get all reviews for a PR | prNumber |
add_task_comment | Add a governed comment with audit trail | issueNumber, comment |
get_sub_issues | Get sub-issues of a parent issue with completion progress | issueNumber |
Container Management (dynamic)
Container tools are generated from the profile’s container definitions. Each managed container type gets its own set of tools.
Execution Container Tools
For the execution container (Wave/Sprint/Cycle):
| Tool Pattern | Hydro | Scrum | Shape Up |
|---|---|---|---|
list_{plural} | list_waves | list_sprints | list_cycles |
get_{singular}_status | get_wave_status | get_sprint_status | get_cycle_status |
create_{singular} | create_wave | create_sprint | create_cycle |
assign_task_to_{singular} | assign_task_to_wave | assign_task_to_sprint | assign_task_to_cycle |
validate_{singular}_completion | validate_wave_completion | validate_sprint_completion | validate_cycle_completion |
Grouping Container Tools
For grouping containers (Epic/Bet):
| Tool Pattern | Hydro | Scrum | Shape Up |
|---|---|---|---|
search_{plural} | search_epics | search_epics | search_bets |
get_{singular}_tasks | get_epic_tasks | get_epic_tasks | get_bet_tasks |
get_{singular}_timeline | get_epic_timeline | get_epic_timeline | get_bet_timeline |
validate_{singular}_integrity | validate_epic_integrity | validate_epic_integrity | validate_bet_integrity |
Shape Up also generates Scope tools (list_scopes, etc.) since Scope is a third container type.
Multi-Agent Coordination (7 tools)
Agent Management
| Tool | Description | Key Parameters |
|---|---|---|
register_agent | Register an AI agent with role and capabilities | agentId, name, role, capabilities? |
list_agents | List all registered agents and their status | — |
lock_task | Acquire exclusive lock (30min TTL) | issueNumber, agentId? |
release_task | Release a task lock | issueNumber, agentId? |
Intelligent Distribution
| Tool | Description | Key Parameters |
|---|---|---|
get_next_task | Scored task recommendation: cascade, momentum, capability match, freshness | agentId?, waveName? |
complete_and_handoff | Atomic: approve + release lock + identify unblocked + recommend next | issueNumber, agentId? |
get_coordination_state | Full multi-agent picture: agents, events, locks, recommendations | agentId?, since? |
Audit, Analytics & Compliance (5 tools)
| Tool | Description | Key Parameters |
|---|---|---|
query_audit_trail | Query events by time, actor, transition, issue, session | since?, until?, actorId?, transition?, issueNumber?, limit? |
get_audit_summary | Event counts grouped by type, actor, and transition | since?, until? |
get_analytics | Velocity, cycle time, lead time, throughput, blocking time | waveName?, since?, until? |
get_task_cycle_time | Full timeline for a specific task | issueNumber |
compute_compliance_score | Deterministic 0-100 score with 5-category breakdown | since?, waveName?, actorId? |
Quality Gate & Dependencies (3 tools)
| Tool | Description | Key Parameters |
|---|---|---|
check_merge_readiness | 6-check gate: workflow, PR review, deps, integrity, security, compliance | issueNumber, overrideReason?, minReviews?, minComplianceScore? |
analyze_dependencies | Full dependency tree: depth, circular detection, blocker identification | issueNumber |
validate_dependencies | Check if all dependencies for a task are satisfied | issueNumber |
Project & Sandbox (5 tools)
| Tool | Description | Key Parameters |
|---|---|---|
init_project | Initialize ido4 governance: create/connect GitHub Project V2, set up fields and config | mode (‘create’ or ‘connect’), repository?, projectName?, methodology? |
get_project_status | Dashboard: container summaries, task distribution, completion metrics | — |
create_sandbox | Create methodology-specific sandbox with embedded violations | repository, methodology? |
destroy_sandbox | Destroy sandbox: close issues, delete project, remove config | — |
reset_sandbox | Destroy and recreate fresh | — |
Ingestion (2 tools)
| Tool | Description | Key Parameters |
|---|---|---|
ingest_spec | Ingest a technical spec: parse, validate, create GitHub issues | specContent, dryRun? |
parse_strategic_spec | Parse a strategic spec from ido4shape into structured data | specContent |
Composite Intelligence (5 tools)
These tools aggregate multiple data sources into single responses, optimized for skill consumption:
| Tool | Replaces | Used By |
|---|---|---|
get_standup_data | 10-12 individual calls (container status, tasks, PRs, deps, audit, analytics, agents, compliance) | /ido4dev:standup |
get_board_data | 5-6 individual calls (container status, tasks with PR/lock annotations, analytics, agents) | /ido4dev:board |
get_compliance_data | 7+ individual calls (compliance score, audit, analytics, containers, tasks, deps) | /ido4dev:compliance |
get_health_data | 5 individual calls (container status, compliance, analytics, agents) | /ido4dev:health |
get_task_execution_data | 5-10 individual calls (task spec, upstream deps with context, siblings, downstream, progress) | Task execution |
Response Format
Every tool returns a consistent structure:
{ "success": true, "data": { }, "suggestions": [ { "action": "review_task", "description": "Task #42 has been in progress for 3 days", "priority": "high" } ], "warnings": [ "Container integrity warning: Auth tasks span 2 waves" ]}