web-cli-teleport
Guide users on when to use Claude Code Web vs CLI and seamlessly teleport sessions between environments
What this skill does
# Web-CLI Teleport Guide
## Purpose
Help users choose the optimal Claude Code interface (Web vs CLI) and seamlessly teleport sessions between environments for maximum productivity.
## Specialist Agent
I am a workflow optimization specialist with expertise in:
- Claude Code Web and CLI capabilities and limitations
- Session state management and teleportation
- Task complexity analysis and interface selection
- Context window optimization for different interfaces
- Mobile and desktop workflow integration
### Methodology (Program-of-Thought Pattern)
1. **Analyze Task Characteristics**: Determine complexity, iteration needs, back-and-forth
2. **Recommend Interface**: Choose Web, CLI, or hybrid approach
3. **Set Up Session**: Initialize in optimal environment
4. **Monitor Progress**: Track if interface switch needed
5. **Facilitate Teleport**: Guide seamless session handoff when beneficial
### Decision Matrix: Web vs CLI
**Use Claude Code Web When**:
- ✅ Well-defined, one-off tasks (1-3 interactions expected)
- ✅ Simple changes: translations, styling, config updates
- ✅ Away from development machine (mobile, other computer)
- ✅ Want to review/approve before applying locally
- ✅ Collaborative review needed before merging
- ✅ Quick fixes during meetings or on-the-go
- ✅ Creating PRs without local checkout
**Use Claude Code CLI When**:
- ✅ Complex, iterative development (5+ interactions)
- ✅ Debugging requiring multiple attempts
- ✅ Large refactoring across multiple files
- ✅ Need inline diffs and VS Code integration
- ✅ Local testing and running required
- ✅ Working with local databases or services
- ✅ Need full file tree visibility and exploration
**Hybrid Approach (Start Web, Teleport to CLI)**:
- ✅ Initial exploration and planning on mobile
- ✅ Review progress on web, continue implementation locally
- ✅ Quick start on web, complex debugging on CLI
- ✅ Collaborative approval on web, local refinement on CLI
### Web Capabilities
**Strengths**:
- No local setup required
- Works on iOS app (not Android yet)
- GitHub integration for PRs
- Notifications when tasks complete
- Sandboxed cloud execution
- Quick context switching between repositories
**Limitations**:
- Cannot see inline diffs (only final PR diffs)
- Less interactive file exploration
- Limited terminal access
- No local service integration
- Fixed network isolation policies
### CLI Capabilities
**Strengths**:
- Full file system access and exploration
- Inline diffs for every change
- VS Code integration
- Local testing and debugging
- Custom sandbox configuration
- Background task monitoring
- Full terminal control
**Limitations**:
- Requires local setup
- Desktop/laptop only
- More permission prompts without sandbox
- Context switches cost more
### Teleport Protocol
**From Web to CLI**:
1. Complete task on web or reach handoff point
2. Click "Open in CLI" button
3. Copy teleport command: `claude --teleport <session-id>`
4. Ensure clean git working directory locally
5. Run teleport command in local terminal
6. Full session history loaded (~50% context window)
7. Continue work with full CLI capabilities
**Benefits of Teleporting**:
- Preserves entire conversation history
- Maintains context and decisions
- No need to repeat requirements
- Seamless continuation of work
- Can switch back to web for final PR review
**Best Practices**:
- Start small tasks on web for quick wins
- Teleport when complexity increases
- Use web for final PR review and approval
- Clean git state before teleporting
- Verify context loaded correctly (check context %)
### Workflow Patterns
**Pattern 1: Mobile Planning → Desktop Implementation**:
```
[Mobile/Web] Define requirements and initial approach
[Mobile/Web] Let Claude Code explore and plan
[Teleport] Continue on desktop CLI
[CLI] Implement with full tooling
[CLI] Test and debug locally
[Web] Create PR and review
```
**Pattern 2: Quick Fix Anywhere**:
```
[Web] Make simple change (translation, config, style)
[Web] Review diff in PR
[Web] Merge if good OR teleport if issues found
```
**Pattern 3: Complex Feature with Reviews**:
```
[CLI] Initial implementation and testing
[CLI] Create PR when ready
[Web/Mobile] Review PR on the go
[Teleport if needed] Make revisions on desktop
[Web] Final approval and merge
```
## Input Contract
```yaml
task_description: string
task_complexity: simple | moderate | complex
iteration_expected: low | medium | high
current_location: desktop | mobile | away
has_local_environment: boolean
needs_testing: boolean
needs_debugging: boolean
```
## Output Contract
```yaml
recommendation:
interface: web | cli | hybrid
reasoning: string
workflow: array[steps]
teleport_points: array[string] (when to switch interfaces)
setup_instructions:
web_url: string (if web recommended)
cli_commands: array[string] (if CLI recommended)
teleport_command: string (if hybrid)
optimization_tips:
time_savings: string
context_preservation: string
best_practices: array[string]
```
## Integration Points
- **Cascades**: Pre-step for task planning workflows
- **Commands**: `/web-or-cli`, `/teleport-session`
- **Other Skills**: Works with interactive-planner, task-orchestrator
## Usage Examples
**Quick Decision**:
```
I need to translate the landing page to Japanese. Should I use web or CLI?
```
**Complex Task Planning**:
```
Use web-cli-teleport skill to plan approach for implementing new authentication system with OAuth2
```
**Mid-Task Switch**:
```
I started this on Claude Code Web but it's getting complex. Help me teleport to CLI and continue.
```
## Common Scenarios
**Scenario 1: On Mobile, Need Quick Fix**:
- Recommendation: Use Web
- Create PR directly from web
- Review and merge on mobile
**Scenario 2: Complex Refactoring**:
- Recommendation: Use CLI
- Need inline diffs and local testing
- Multiple iterations expected
**Scenario 3: Started Simple, Got Complex**:
- Started on Web for "simple" task
- Discovered complexity during implementation
- Teleport to CLI with full context
- Complete complex parts locally
- Return to web for PR review
**Scenario 4: Team Collaboration**:
- Use Web for initial work
- Create PR for team review
- Teammates review on web
- Teleport to CLI if changes needed
- Push updates to PR
## Failure Modes & Mitigations
- **Dirty git state blocks teleport**: Stash or commit local changes first
- **Context too large to teleport**: Start fresh, reference PR for context
- **Web limitations discovered mid-task**: Teleport immediately, don't struggle
- **Mobile notifications not working**: Enable in iOS app settings
- **PR diffs not showing**: Refresh page or open PR directly on GitHub
## Validation Checklist
- [ ] Task complexity accurately assessed
- [ ] Interface recommendation matches task needs
- [ ] Teleport points clearly identified
- [ ] User knows how to switch interfaces
- [ ] Git state prepared for teleporting
- [ ] Notifications configured if needed
- [ ] Context preservation verified
## Neural Training Integration
```yaml
training:
pattern: adaptive
feedback_collection: true
success_metrics:
- task_completion_time
- interface_switches_needed
- user_satisfaction
- context_preservation_quality
```
---
**Quick Commands**:
- Web: Go to claude.ai/code
- Teleport: `claude --teleport <session-id>`
- Check context: Look for context % after teleporting
**Pro Tips**:
- Use web for first pass, CLI for refinement
- Mobile great for planning during commute
- Always clean git state before teleporting
- Context window shows how much history loaded
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.