Full-power feature implementation using parallel subagents for backend, frontend, testing, and security. Coordinates architecture design, code generation, test coverage, and quality verification in a single workflow with worktree isolation. Chains with /ork:cover for test generation and /ork:verify for validation. Use when implementing features, building new capabilities, or creating full-stack functionality.
Full-power feature implementation using parallel subagents for backend, frontend, testing, and security. Coordinates architecture design, code generation, test coverage, and quality verification in a single workflow with worktree isolation. Chains with /ork:cover for test generation and /ork:verify for validation. Use when implementing features, building new capabilities, or creating full-stack functionality.
{"keywords":["implement","implment","build","create","add","make","scaffold","set up","file upload","dark mode","rate limiting"],"examples":["build a user authentication system with JWT","add dark mode support to the dashboard","implement the payment webhook handler"],"anti-triggers":["fix","debug","review","explore","test","assess","brainstorm"]}
Parallel subagent execution for feature implementation with scope control and reflection.
Quick Start
/ork:implement user authentication
/ork:implement --model=opus real-time notifications
/ork:implement dashboard analytics
Argument Resolution
FEATURE_DESC = "$ARGUMENTS"# Full argument string, e.g., "user authentication"# $ARGUMENTS[0] is the first token, $ARGUMENTS[1] second, etc. (CC 2.1.59)# Model override detection (CC 2.1.72)
MODEL_OVERRIDE = Nonefor token in"$ARGUMENTS".split():
if token.startswith("--model="):
MODEL_OVERRIDE = token.split("=", 1)[1] # "opus", "sonnet", "haiku"
FEATURE_DESC = FEATURE_DESC.replace(token, "").strip()
Pass MODEL_OVERRIDE to all Agent() calls via model=MODEL_OVERRIDE when set. Accepts symbolic names (opus, sonnet, haiku) or full IDs (claude-opus-4-6) per CC 2.1.74.
Step -1: MCP Probe + Resume Check
Run BEFORE any other step. Detect available MCP servers and check for resumable state.
# Probe MCPs (parallel — all in ONE message):
ToolSearch(query="select:mcp__memory__search_nodes")
ToolSearch(query="select:mcp__context7__resolve-library-id")
Write(".claude/chain/capabilities.json", JSON.stringify({
"memory": <true if found>,
"context7": <true if found>,
"timestamp": now()
}))
# Resume check:
Read(".claude/chain/state.json")
# If exists and skill == "implement":
Write(, JSON.stringify({
: , : FEATURE_DESC,
: , : [],
: capabilities,
: // advisory; see Budget Awareness below
}))
# Read last handoff (e.g., 04-architecture.json)
# Skip to current_phase
# "Resuming from Phase {N} — architecture decided in previous session"
# If not: write initial state
".claude/chain/state.json"
"skill"
"implement"
"feature"
"current_phase"
1
"completed_phases"
"capabilities"
"budget_remaining_pct"
100
Batch Size Governance (large refactors)
For implementations touching >10 files, enforce max 5 files per agent batch, run tests between batches, commit green batches immediately, stop on red. Override via --batch-size N. Full rule: Read("${CLAUDE_SKILL_DIR}/rules/batch-governance.md").
Budget Awareness (Opus 4.7 task budgets, public beta)
Opus 4.7 exposes per-task token budgets. Until the CC side is GA, OrchestKit tracks an advisory budget_remaining_pct in state.json so long runs self-throttle. Update after each phase:
# At end of every phase, estimate remaining budget:
pct = tokensAsContextPct(tokensUsedSoFar) # from lib/context-window.ts
remaining = max(0, 100 - pct)
state["budget_remaining_pct"] = remaining
Write(".claude/chain/state.json", JSON.stringify(state))
Thresholds influence behavior:
Remaining
Behavior
> 50%
Normal — all optional depth (devil's advocate, visual capture, deep exploration).
Read the /effort setting to scale implementation depth. The effort-aware context budgeting hook detects effort level automatically — adapt the phase plan accordingly:
Effort Level
Phases Run
Agents
Token Budget
low
1 (Discovery) → 5 (Implement) → 10 (Reflect)
2 max
~50K
medium
1 → 2 → 5 → 7 (Scope Creep) → 10
3 max
~150K
high (default)
All 10 phases
4-7
~400K
xhigh (Opus 4.7 only, CC 2.1.111+)
All 10 phases + one additional healing iteration on test failures before escalating
4-7
~550K
Override: Explicit user selection in Step 0 (e.g., "Plan first" or "Worktree") overrides /effort downscaling. If user requests full exploration, respect that regardless of effort level.
Step 0a: Project Context Discovery
BEFORE any work, detect the project tier. This becomes the complexity ceiling for all patterns.
Scan codebase signals and classify into tiers 1-6 (Interview through Open Source). Each tier sets an architecture ceiling and determines which phases/agents to use.
Load tier details, workflow mapping, and orchestration mode: Read("${CLAUDE_SKILL_DIR}/references/tier-classification.md")
Worktree Isolation (CC 2.1.49)
For features touching 5+ files, offer worktree isolation to prevent conflicts with the main working tree:
AskUserQuestion(questions=[{
"question": "Isolate this feature in a git worktree?",
"header": "Isolation",
"options": [
{"label": "Yes — worktree (Recommended)", "description": "Creates isolated branch via EnterWorktree, merges back on completion", "markdown": "```\nWorktree Isolation\n──────────────────\nmain ─────────────────────────────▶\n \\ /\n └─ feat-{slug} (worktree) ───┘\n ├── Isolated directory\n ├── Own branch + index\n └── Auto-merge on completion\n\nSafe: main stays untouched until done\n```"},
{"label": "No — work in-place", "description": "Edit files directly in current branch", "markdown": "```\nIn-Place Editing\n────────────────\nmain ──[edit]──[edit]──[edit]───▶\n ▲ ▲ ▲\n │ │ │\n direct modifications\n\nFast: no branch overhead\nRisk: changes visible immediately\n```"},
{"label": "Plan first", "description": "Research and design in plan mode before writing code", "markdown": "```\nPlan Mode Flow\n──────────────\n 1. EnterPlanMode($ARGUMENTS)\n 2. Read existing code\n 3. Research patterns\n 4. Design approach\n 5. ExitPlanMode → plan\n 6. User approves plan\n 7. Execute implementation\n\n Best for: Large features,\n unfamiliar codebases,\n architectural decisions\n```"}
],
"multiSelect": false
}])
If 'Plan first' selected:
# 1. Enter read-only plan mode
EnterPlanMode("Research and design: $ARGUMENTS")
# 2. Research phase — Read/Grep/Glob ONLY, no Write/Edit# - Read existing code in the target area# - Grep for related patterns, imports, dependencies# - Check tests, configs, and integration points# - If context7 available: query library docs# 3. Design the plan — produce:# - File map: which files to create/modify# - Architecture decisions with rationale# - Task breakdown with acceptance criteria# - Risk assessment and edge cases# 4. Exit plan mode — returns plan to user for approval
ExitPlanMode()
# 5. User reviews plan. If approved → continue to Phase 1 (Discovery)# with the plan as input. If rejected → revise or stop.
If worktree selected:
Call EnterWorktree(name: "feat-{slug}") to create isolated branch
For Agent Teams mode: Read("${CLAUDE_SKILL_DIR}/references/agent-teams-phases.md")
Phase Handoffs (CC 2.1.71)
Write handoff JSON after major phases. See chain-patterns skill for schema.
After Phase
Handoff File
Key Outputs
1. Discovery
01-discovery.json
Best practices, library docs, task breakdown
2. Micro-Plan
02-plan.json
File map, acceptance criteria per task
4. Architecture
04-architecture.json
Decisions, patterns chosen, agent results
5. Implementation
05-implementation.json
Files created/modified, test results
7. Scope Creep
07-scope.json
Planned vs actual, PR split recommendation
Progressive Output (CC 2.1.76+)
Output results incrementally after each phase — don't batch everything until the end.
Focus mode (CC 2.1.101): In focus mode (/focus), the user only sees your final message. Include a self-contained summary with all key results — don't assume they saw incremental outputs.
When agents run with run_in_background=true, output each agent's findings as soon as it returns — don't wait for all agents to finish. This gives users ~60% faster perceived feedback and enables early intervention if an agent's approach diverges from the plan.
Monitor Tool for Background Streaming (CC 2.1.98)
Use Monitor to stream real-time events from background build/test scripts instead of polling output files:
# Start a long-running build in background
Bash(command="npm run build 2>&1", run_in_background=true)
# Stream its output line-by-line as notifications
Monitor(pid=build_task_id)
# Each stdout line arrives as a notification — no polling needed# For background agents with test suites:
Agent(subagent_type="test-generator", run_in_background=true, ...)
# Monitor agent progress via task notifications (CC 2.1.98 partial progress)
Full pattern reference (when to use vs. TaskOutput, until-condition gates, partial-result salvage, anti-patterns): Read("/Users/yonatangross/coding/yonatangross/orchestkit/plugins/ork/skills/chain-patterns/references/monitor-patterns.md").
Partial results (CC 2.1.98): Background agents that fail now report partial progress to the parent. If a worktree-isolated agent crashes mid-implementation, synthesize its partial output instead of re-spawning:
# After collecting agent results:for agent_result in agent_results:
if"[PARTIAL RESULT]"in agent_result.output:
# Agent crashed mid-work — salvage what it produced
partial_files = Bash(command="git diff --name-only", cwd=agent_result.worktree)
if partial_files:
# Merge partial work — commit what's usable, flag incomplete items
TaskUpdate(taskId=agent_task_id, status="completed",
description=f"Partial: {len(partial_files)} files from crashed agent")
# Do NOT re-spawn — partial progress > wasted tokens re-doing workelif agent_result.status == "BLOCKED":
# Agent hit a genuine blocker — escalate to user
TaskUpdate(taskId=agent_task_id, status="in_progress",
description=f"BLOCKED: {agent_result.concerns[0]}")
Worktree-Isolated Implementation (CC 2.1.50)
Phase 5 agents SHOULD use isolation: "worktree" to prevent file conflicts:
# Guard: Skip cron in headless/CI (CLAUDE_CODE_DISABLE_CRON)# if env CLAUDE_CODE_DISABLE_CRON is set, run a single check instead
CronCreate(
schedule="0 */6 * * *",
prompt="Health check for {feature} in PR #{pr}:
gh pr checks {pr} --repo {repo}.
If healthy 24h → CronDelete. If errors → alert."
)
If working on a GitHub issue, run the Start Work ceremony from issue-progress-tracking and post progress comments after major phases.
Feedback Loop
Maintain checkpoints after each task. Load triggers: Read("${CLAUDE_SKILL_DIR}/references/feedback-loop.md")
Test Requirements Matrix
Phase 5 test-generator MUST produce tests matching the change type. Each change type maps to specific required tests and testing rules.
Load test matrix, real-service detection, and phase 9 gate: Read("${CLAUDE_SKILL_DIR}/references/test-requirements-matrix.md")
Key Principles
Verification gate — before claiming ANY task done, apply the 5-step gate: Read("${CLAUDE_PLUGIN_ROOT}/skills/shared/rules/verification-gate.md"). "Should work now" is not evidence.
Agent status protocol — all subagents report DONE / DONE_WITH_CONCERNS / BLOCKED / NEEDS_CONTEXT per Read("${CLAUDE_PLUGIN_ROOT}/agents/shared/status-protocol.md")
Tests are NOT optional — each task includes its tests, matched to change type (see matrix above)
Parallel when independent — use run_in_background: true, launch all agents in ONE message
Output limits (CC 2.1.77+): Opus 4.6 defaults to 64k output tokens (128k upper bound). Generate complete artifacts in a single pass when possible; chunk across turns if output exceeds the limit
Micro-plan before implementing — scope boundaries, file list, acceptance criteria
Real services when available — if docker-compose/testcontainers exist, use them in Phase 6
Reflect and capture lessons (phase 10) — persist to memory graph
Clean up agents — use TeamDelete() after completion; press Ctrl+F twice as manual fallback. Note: /clear (CC 2.1.72+) preserves background agents
Exit worktrees — call ExitWorktree(action: "keep") in Phase 10 if worktree was entered in Step 0; never leave orphaned worktrees
Next Steps (suggest to user after implementation)
/ork:verify {FEATURE} # Grade the implementation
/ork:cover {FEATURE} # Generate test suite
/ork:commit # Commit changes
/loop 10m npm test # Watch tests while iterating
/loop 30m /ork:verify {FEATURE} # Periodic quality gate
PushNotification on Completion (CC 2.1.110+)
/ork:implement runs commonly take 10–30 min with parallel agents. At the final synthesis step, after the PR is opened and tests are green, call PushNotification — the user has almost certainly context-switched.
PushNotification(
title="ork:implement complete",
body=f"{FEATURE}: {tests_passing}/{tests_total} tests · PR #{pr_num} opened · ready for /ork:verify"
)
Full rule (when to fire, body content limits, graceful fallback for users without Remote Control): load Read("/Users/yonatangross/coding/yonatangross/orchestkit/plugins/ork/skills/chain-patterns/rules/push-notification-on-completion.md").
Agent Coordination
Context Passing
All spawned agents receive: changed files list, project tier, architectural constraints, and decisions from prior phases (discovery, plan). Pass via the agent prompt, not just "implement X".
SendMessage (Active Coordination)
When backend and frontend agents need to align on API contracts:
SendMessage(to="frontend-ui-developer", message="API endpoint is POST /api/auth with {token, refreshToken} response shape")
SendMessage(to="test-generator", message="Backend uses JWT — mock auth middleware in test fixtures")
Skill Chain
After implementation completes, chain to verification:
Session recovery (CC 2.1.108+): After idle periods or interruptions, use /recap to restore conversational context. Combined with .claude/chain/state.json checkpoint-resume, this enables full recovery of multi-phase implement sessions. Enabled by default since CC 2.1.110 (even with telemetry disabled).
Related Skills
ork:explore: Explore codebase before implementing
ork:verify: Verify implementations work correctly
ork:issue-progress-tracking: Auto-updates GitHub issues with commit progress
References
Load on demand with Read("${CLAUDE_SKILL_DIR}/references/<file>"):
