VGCHelper
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| VGC_HELPER_DATA_DIR | No | Override the local state directory. By default, local state is written to `.vgc-helper\vgc-helper.sqlite` relative to the server's working 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 | {
"listChanged": true
} |
| prompts | {
"listChanged": true
} |
| resources | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| vgc_statusA | Show the active regulation, calculator version, local database, and metagame source freshness. |
| vgc_refresh_metaB | Fetch, validate, cache, and atomically activate VGC Pastes teams and Champions Battle Data usage for the active regulation. |
| vgc_replay_analyzeB | Parse and persist a Champions VGC doubles replay, then return turn-cited coaching evidence using the supplied exact user team. |
| vgc_replay_getA | Retrieve a persisted replay coaching report by analysis ID. |
| vgc_replay_turnA | Retrieve stored state before and after a turn plus battle events. Judge decisions using beforeEvents only; afterEvents contains later revelations. |
| vgc_meta_queryA | Query cached team sets and per-Pokemon usage with source dates and regulation verification. Unverified usage is contextual only. |
| vgc_replay_trendsB | Aggregate recurring strengths, improvement findings, and results across locally persisted replay analyses. |
| vgc_team_evaluateB | Evaluate legal Mega alternatives and fixed matchup modes against a threat-aware cohort. Optionally compare an edited team on identical sources and conditions; scores are pressure heuristics. |
| vgc_matchup_detailB | Retrieve bounded lead-state details from a persisted team evaluation. |
| vgc_damage_calculateA | Run an auditable generation-zero Pokemon Champions doubles damage calculation from two Showdown-format sets. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| replay-coach | Analyze a replay with evidence-first coaching and explicit hidden-information uncertainty. |
| team-builder | Interpret a saved lead-matrix evaluation without presenting heuristic scores as win rates. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| active-regulation | The immutable regulation profile used by default. |
| methodology | Analysis boundaries and interpretation rules. |
| source-status | Active source versions and retrieval times. |
TDQS
Scored across 10 tools
Each tool targets a distinct resource or action: status, meta refresh/query, replay analysis/retrieval/turn inspection/trends, team evaluation/detail, and damage calculation. There is no meaningful overlap among the tools even though several operate on replays or evaluations.
All tools share the vgc_ prefix and use lowercase snake_case, but the verb/noun ordering is inconsistent: refresh_meta and meta_query flip the pattern, while replay_analyze, replay_get, team_evaluate, and damage_calculate place the object before the verb. The names remain readable, but the convention is not uniform.
With 10 tools, the server is well-scoped for a VGC helper covering metagame data, replay coaching, team evaluation, and damage calculation. Each tool addresses a meaningful part of the workflow without unnecessary redundancy.
The tool surface covers the main lifecycle needs: refreshing and querying meta data, analyzing and retrieving replay coaching details, aggregating trends, evaluating teams, and calculating damage. Minor gaps exist, such as no listing or deletion of persisted replay reports/evaluations, but these are workable and do not break core workflows.