VibeTape MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| VIBETAPE_HOME | No | Optional: Custom storage location (default ~/.vibetape) | |
| OPENAI_API_KEY | No | Optional: OpenAI API key for semantic search and RETEX generation | |
| VIBETAPE_TEAM_DIR | No | Optional: Team collaboration directory |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
| prompts | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| mark_momentC | Capture a build moment with safe context snapshot |
| link_momentsB | Create a relation between moments |
| comment_momentB | Append a comment to a moment |
| list_momentsA | Return latest captured moments |
| search_momentsC | Find similar moments using semantic search |
| search_moments_advancedC | Combine semantic, filters and regex search |
| stats_overviewD | Basic counts and trends |
| recurrent_patternsB | Group similar titles to find patterns |
| make_retexC | Generate rule + bullets + dont from a captured moment |
| export_timelineB | Render Markdown timeline for a given day (YYYY-MM-DD) |
| export_jsonB | Return whole state as JSON resource |
| export_mdD | Return Markdown dump resource |
| generate_context_handoffC | Create a compact transition card (state/stack/decisions/solved/next) under a token budget |
| suggest_transition_cardB | Suggest generating a transition card when context is near capacity |
| sweep_noiseB | Denoise auto-marked moments: trivial, duplicates, cooldown; updates signal_score |
| register_actorB | Register a new agent or human actor for traceability |
| get_actor_statsA | Get statistics for an actor (success rate, moment counts, top tags) |
| list_actorsA | List all registered actors |
| create_taskB | Create a new task for tracking and agent assignment |
| update_taskA | Update a task status, assignment, or outcome |
| list_tasksC | List tasks with optional filters |
| assign_taskC | Assign or handoff a task to another actor |
| link_moment_to_taskC | Associate a moment with a task |
| supersede_momentC | Mark a moment as superseded by a newer one (temporal tracking) |
| get_retex_for_taskA | Get relevant RETEX cards for a task based on tag overlap and moment relations |
| create_handoff_for_agentC | Create a LangGraph-compatible handoff payload for agent-to-agent transfer |
| context_relevance_scoreC | Calculate RankRAG-style relevance score for a moment given a task context |
| evaluate_context_windowC | Evaluate and rank moments for optimal context injection within token budget |
| predict_agent_needsB | Predict what context an agent will need based on task type and actor capabilities |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| induce-retex | Induce RETEX from moment text |
| commit-msg | Conventional commit helper |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Build moment | Captured key build moment (JSON) |
| Timeline Markdown | Day timeline from captured moments |
| RETEX card | Induced prescriptive card (JSON) |
| Moment subgraph | Relations for a moment |
| Context Handoff | Compact transition card for session continuity |
| JSON export | Filtered JSON export |
| Markdown export | Markdown dump of moments |
| Actor profile | Actor with stats (JSON) |
| Task details | Task with related moments (JSON) |
| Task context | Optimized context for task injection (Markdown) |
| VibeTape capabilities | List of VibeTape features for agent discovery (JSON) |
TDQS
Scored across 29 tools
Most tools target distinct resource-action pairs, e.g., moment capture vs. task management. The main potential confusion is between generate_context_handoff and create_handoff_for_agent, but they produce different output formats (human-readable card vs. LangGraph payload).
Tool names are uniformly snake_case and mostly follow verb_noun (export_timeline, create_task). A few names like context_relevance_score, stats_overview, and recurrent_patterns deviate from the verb-first pattern, but the overall style is consistent and readable.
With 29 tools, the surface is quite large and exceeds the typical well-scoped range. While the broad domain (moments, tasks, actors, handoffs, analytics) somewhat justifies the count, it still feels heavy and likely overwhelming for an agent.
The toolset covers the main lifecycle for moments (capture, search, comment, link, supersede), tasks (create, update, list, assign), and actors (register, list, stats). Minor gaps include lack of delete operations for moments/tasks/actors, but core workflows are functional.