Codebase Exploration
Multi-angle codebase exploration using 3-5 parallel agents.
๐ฏ Quick Start
/ork:explore authentication
Opus 5: Exploration agents use native adaptive thinking for deeper pattern recognition across large codebases.
STEP -0.5: Effort-Aware Agent Scaling (CC 2.1.120+)
Read $CLAUDE_EFFORT to scale exploration depth before any other decision.
EFFORT = os.environ.get("CLAUDE_EFFORT")
for token in "$ARGUMENTS".split():
if token.startswith("--effort="):
EFFORT = token.split("=", 1)[1]
EFFORT = EFFORT or "high"
| Effort | Agent count | Phases | Time |
|---|
low | 1 (structure-only) | 1, 2, 8 | ~1 min |
medium | 2 (structure + data flow) | 1, 2, 3 (subset), 8 | ~3 min |
high (default) | 4 (full parallel team) | 1โ8 | ~6 min |
xhigh (Opus 5) | 5 (+ uncertainty pass on health scores) | 1โ8 + caveats | ~8 min |
Override gate: if the user passes --effort=high explicitly while $CLAUDE_EFFORT is low, the flag wins. /ork:doctor warns when xhigh is requested without Opus 5.
STEP 0: Verify User Intent with AskUserQuestion
BEFORE creating tasks, clarify what the user wants to explore:
AskUserQuestion(
questions=[{
"question": "What aspect do you want to explore?",
"header": "Focus",
"options": [
{"label": "Full exploration (Recommended)", "description": "Code structure + data flow + architecture + health assessment"},
{"label": "Quick scan", "description": "Find relevant files + structure, skip deep analysis"},
{"label": "Data flow", "description": "Trace how data moves through the system"},
{"label": "Architecture patterns", "description": "Identify design patterns and integrations"}
],
"multiSelect": false
}]
)
Based on answer, adjust workflow:
- Full exploration: All phases, all parallel agents
- Quick scan: Files + structure only (phases 1-2), skip health/deps/product โ no deep agents
- Data flow: Focus phase 3 agents on data tracing
- Architecture patterns: Focus on backend-system-architect agent
STEP 0b: Select Orchestration Mode
MCP Probe
ToolSearch(query="select:mcp__memory__search_nodes")
Write(".claude/chain/capabilities.json", { memory, timestamp })
if capabilities.memory:
mcp__memory__search_nodes({ query: "architecture decisions for {path}" })
Exploration Handoff
After exploration completes, write results for downstream skills:
Write(".claude/chain/exploration.json", JSON.stringify({
"phase": "explore", "skill": "explore",
"timestamp": now(), "status": "completed",
"outputs": {
"architecture_map": { ... },
"patterns_found": ["repository", "service-layer"],
"complexity_hotspots": ["src/auth/", "src/payments/"]
}
}))
Choose Agent Teams (mesh) or Task tool (star):
- Agent Teams mode (GA since CC 2.1.33) โ recommended for 4+ agents
- Task tool mode โ for quick/single-focus exploration
ORCHESTKIT_FORCE_TASK_TOOL=1 โ Task tool (override)
| Aspect | Task Tool | Agent Teams |
|---|
| Discovery sharing | Lead synthesizes after all complete | Explorers share discoveries as they go |
| Cross-referencing | Lead connects dots | Data flow explorer alerts architecture explorer |
| Cost | ~150K tokens | ~400K tokens |
| Best for | Quick/focused searches | Deep full-codebase exploration |
Fallback: If Agent Teams encounters issues, fall back to Task tool for remaining exploration.
Model cost (CC 2.1.198+): the built-in Explore agent inherits the session model capped at Opus โ it no longer runs on haiku. From a premium-model session (Opus, Fable), budget Explore fan-outs at Opus rates; there is no knob to pin the built-in Explore back to haiku. ork's own explorer agents can still pin a cheaper model via frontmatter.
๐จ Task Management (MANDATORY)
BEFORE doing ANYTHING else, create tasks to show progress:
TaskCreate(subject="Explore: {topic}", description="Deep codebase exploration for {topic}", activeForm="Exploring {topic}")
TaskCreate(subject="Initial file search", activeForm="Searching files")
TaskCreate(subject="Check knowledge graph", activeForm="Checking memory")
TaskCreate(subject="Launch exploration agents", activeForm="Dispatching explorers")
TaskCreate(subject="Assess code health (0-10)", activeForm="Assessing code health")
TaskCreate(subject="Map dependency hotspots", activeForm="Mapping dependencies")
TaskCreate(subject="Add product perspective", activeForm="Adding product context")
TaskCreate(subject="Generate exploration report", activeForm="Generating report")
TaskUpdate(taskId="3", addBlockedBy=["2"])
TaskUpdate(taskId="4", addBlockedBy=["3"])
TaskUpdate(taskId="5", addBlockedBy=["4"])
TaskUpdate(taskId="6", addBlockedBy=["4"])
TaskUpdate(taskId="7", addBlockedBy=["4"])
TaskUpdate(taskId=, addBlockedBy=[, , ])
TaskUpdate(taskId=, status=)
TaskUpdate(taskId=, status=)
๐ Workflow Overview
| Phase | Activities | Output |
|---|
| 1. Initial Search | Grep, Glob for matches | File locations |
| 2. Memory Check | Search knowledge graph | Prior context |
| 3. Deep Exploration | 4 parallel explorers | Multi-angle analysis |
| 4. AI System (if applicable) | LangGraph, prompts, RAG | AI-specific findings |
| 5. Code Health | Rate code 0-10 | Quality scores |
| 6. Dependency Hotspots | Identify coupling | Hotspot visualization |
| 7. Product Perspective | Business context | Findability suggestions |
| 8. Report Generation | Compile findings | Actionable report |
Progressive Output (CC 2.1.76)
Output findings incrementally as each phase completes โ don't batch until the report:
| After Phase | Show User |
|---|
| 1. Initial Search | File matches, grep results |
| 2. Memory Check | Prior decisions and relevant context |
| 3. Deep Exploration | Each explorer agent's findings as they return |
| 5. Code Health | Health score with dimension breakdown |
For Phase 3 parallel agents, output each agent's findings as soon as it returns โ don't wait for all 4 explorers. Early findings from one agent may answer the user's question before remaining agents complete, allowing early termination.
Phase 1: Initial Search
Grep(pattern="$ARGUMENTS[0]", output_mode="files_with_matches")
Glob(pattern="**/*$ARGUMENTS[0]*")
Phase 2: Memory Check
mcp__memory__search_nodes(query="$ARGUMENTS[0]")
mcp__memory__search_nodes(query="architecture")
Phase 3: Parallel Deep Exploration (4 Agents)
Load Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/rules/exploration-agents.md") for Task tool mode prompts.
Load Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/rules/agent-teams-mode.md") for Agent Teams alternative.
Phase 4: AI System Exploration (If Applicable)
For AI/ML topics, add exploration of: LangGraph workflows, prompt templates, RAG pipeline, caching strategies.
Phase 5: Code Health Assessment
Load Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/rules/code-health-assessment.md") for agent prompt. Load Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/references/code-health-rubric.md") for scoring criteria.
Phase 6: Dependency Hotspot Map
Load Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/rules/dependency-hotspot-analysis.md") for agent prompt. Load Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/references/dependency-analysis.md") for metrics.
Phase 7: Product Perspective
Load Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/rules/product-perspective.md") for agent prompt. Load Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/references/findability-patterns.md") for best practices.
Phase 8: Generate Report
Load Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/references/exploration-report-template.md").
Phase 8b: Emit Dashboard Spec (json-render)
Parse --render= from $ARGUMENTS. Default is both.
| Mode | Behavior |
|---|
markdown | Current behavior โ markdown report only. No spec emitted. |
json-render | Emit .claude/chain/explore-dashboard.json only. Skip markdown report. |
both | Emit spec and markdown. Default โ gives the human a report and downstream skills a structured handoff. |
When emitting a spec:
- Load the format and catalog:
Read("${CLAUDE_PLUGIN_ROOT}/skills/explore/references/dashboard-spec.md"). Reference example: references/dashboard-example.json.
- Build the spec object using only catalog component types:
Card, StatGrid, DataTable, StatusBadge, BarMeter, Heatmap, Markdown.
- Write to
.claude/chain/explore-dashboard.json with compact JSON (no indentation) โ minimizes token cost for downstream consumers.
- Validate before declaring success:
node "${CLAUDE_PLUGIN_ROOT}/skills/explore/scripts/render-spec.mjs" .claude/chain/explore-dashboard.json --check
If validation fails (exit โ 0), do not emit โ fall back to markdown-only and surface the error to the user. Never write a partial or invalid spec.
- For
--render=both, render the markdown view from the spec for consistency:
node "${CLAUDE_PLUGIN_ROOT}/skills/explore/scripts/render-spec.mjs" .claude/chain/explore-dashboard.json
Pipe the output into the user-facing markdown report (or use it as-is). This guarantees the JSON spec and markdown report stay in sync โ a single source of truth.
Why this matters: Downstream skills (/ork:fix-issue, /ork:implement, /ork:create-pr) parse .claude/chain/explore-dashboard.json directly instead of re-reading 3000-token markdown. Measured: spec โ 580 tokens for the same content. Backwards-compatible: old chained workflows that read markdown keep working in both mode.
Phase 6.5 โ Notebook summary (signal-fired, optional)
After the session synthesis lands, optionally invoke scripts/post_explore_summary.py <session-dir> to auto-emit a notebook-backed summary of the exploration. Self-skips on every non-happy-path so it never breaks the run:
python3 ${CLAUDE_PLUGIN_ROOT}/skills/explore/scripts/post_explore_summary.py "$CLAUDE_JOB_DIR"
Auto-skip conditions (all exit 0, all WARN-logged):
| Skip reason | Trigger |
|---|
signal absent | len(dirs_scanned) < 3 (or field missing on explore-output.json) |
yg-mcp-core not importable | yg-mcp-core>=0.3.0 not installed (orchestkit is public; yg-mcp-core lives on private pypi.yonyon.ai โ HQ-only) |
hq-content MCP unreachable | MCP server down OR .mcp.json doesn't define hq-content |
Session dir must contain explore-output.json (with dirs_scanned: list[str], optional synthesis: str, required notebook_id: str). Handoff JSON at <session-dir>/explore-summary.json records status (fired / skipped) and summary_path on success.
Mirrors the /ork:brainstorm post-synth podcast pattern from PR #1889. Closes orchestkit#1893.
Notes for long explorations
Oversized reads (CC 2.1.144+): Read returns a [PARTIAL view] truncated first page (not a hard error) when a whole-file read exceeds the token limit. When traversing large files, detect that notice and re-read with explicit offset/limit to page through the rest โ never treat the partial as the full file.
When context fills (CC 2.1.141+): Use the rewind menu's "Summarize up to here" to compress earlier turns while keeping recent context, instead of restarting. Reactive compaction (CC 2.1.142+) now sizes the first summarize to the actual overflow, so a second mid-turn pass is rare.
Common Exploration Queries
- "How does authentication work?"
- "Where are API endpoints defined?"
- "Find all usages of EventBroadcaster"
- "What's the workflow for content analysis?"
Running unattended with /goal
Set a completion condition with /goal (CC 2.1.139+) and this skill will keep working across turns until the condition is met. Works in interactive, -p, and Remote Control. The overlay panel shows live elapsed / turns / tokens.
Example completion condition for this skill:
/goal until report.has_architecture_diagram AND patterns.detected_count >= 5, or stop after 10 turns
Stops when: codebase architecture diagram is generated and at least 5 design patterns have been classified. Compatible with claude.ai Remote Control runs.
Quality Bar
Done means all of these hold:
- Every architectural or data-flow claim cites concrete evidence (file:line or a file path), not prose assertion
- Code health is reported as 0-10 scores with a per-dimension breakdown, not a bare number
- Dependency hotspots / coupling are named along with the files that drive them
- The report includes an architecture or structure visualization for the explored scope
- If a json-render spec is emitted, it passes
render-spec.mjs --check; on failure fall back to markdown-only and never write a partial spec
๐ Related Skills
ork:implement: Implement after exploration
Version: 2.6.0 (April 2026) โ $CLAUDE_EFFORT env var scales agent count (CC 2.1.120, #1540)
