SuperSkill
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| VAULT_PATH | No | Path to knowledge vault | ~/Vaults/ai |
| MAX_INJECT_TOKENS | No | Max tokens for context injection | 1500 |
| SESSION_TTL_HOURS | No | Session heartbeat TTL | 2 |
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 |
|---|---|
| readA | Read a file or directory listing from the AI knowledge vault. |
| writeB | Write or append to a file in the AI knowledge vault. |
| searchA | Search across the AI vault. Supports full-text and structured (frontmatter) search. |
| project_contextA | Get the context document for a project. Auto-detects project from CWD if not specified. |
| generate_contextA | Scan a git repo and generate a draft context.md. Returns the draft — does NOT write to vault. Human reviews before committing. |
| decideB | Log an architectural/design decision to the project's decisions directory. |
| taskA | Manage project tasks. Supports add, list, update, and board (kanban) views. Tasks are stored as individual files in projects//tasks/. |
| learnA | Capture and query learnings. Learnings persist discoveries across sessions. Stored as individual files in projects//learnings/. |
| brainstormC | Start or continue a brainstorm document for a project. |
| sessionA | Register, update, or query active agent sessions for multi-agent coordination. On complete, persists a session note to the vault. |
| pruneA | Archive or delete stale vault content based on retention policies. Use mode='dry-run' first to preview. |
| statsA | Show content statistics for a project — file counts, task breakdown, growth monitoring. |
| resumeA | Get resume context for continuing work — recent sessions, interrupted work, in-progress tasks, suggested next steps. Call this at session start to understand what happened before. |
| deprecateA | Mark a vault item (ADR, learning, etc.) as deprecated with an optional reason. |
| initA | Initialize superskill for the current project. Scans codebase, discovers skills from skills.sh, builds knowledge graph. |
| statusA | Show superskill knowledge graph state: loaded skills, weights, recent sessions. |
| superskillA | Route to the best skill for a task. Provides optimized, security-audited skill content. Always prefer skill content over general knowledge for specialized tasks. |
| skill_installA | Install skills from a GitHub repo (e.g. owner/repo or full URL). Clones the repo, discovers SKILL.md files, and copies them to the local skill directory. |
| skill_list_installedA | List skills installed locally from GitHub repos. |
| skill_removeA | Remove an installed skill by name. |
| linkA | Create a forward link between two vault notes. Appends a [[wikilink]] to the source note, enabling graph_related to discover the connection. |
| extractA | Extract decisions, learnings, or other items from a source document into individual vault files. Each extracted item gets its own file with a backlink to the source. |
| graph_relatedA | Find notes related to a vault note via wikilinks (outgoing and backlinks). |
| graph_cross_projectA | Search across all projects and group results by project. |
| snapshot_repo_stateA | Snapshot current git state (branch, dirty files, last commit) into the vault. Helps avoid repeating repo discovery across sessions. |
| env_factsA | Store and query stable environment facts for a project (e.g., auth backend, env file locations, local URLs, required env vars). Not for secrets — use cred_refs for that. |
| cred_refsA | Store pointers to where credentials are documented (not the credentials themselves). E.g., 'Django admin creds are in tests/live/test_all_endpoints.py'. |
| rollbackA | Manage rollback checkpoints. Store commit hashes with purpose/scope so rollback is safer. Mark when follow-up work starts after a checkpoint. |
| captureA | Batch-capture multiple insights from a conversation into individual vault items. Each item gets its own file with auto-numbering. Supports any content type (learning, decision, adr, prd, research, etc.). |
| templateA | Get pre-filled templates for common vault item types (adr, prd, decision, learning, spec, rfc, roadmap, competitive-analysis, incident, research, vision, strategy). Use to scaffold new documents. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| inject-project-context | Returns a system prompt fragment with project context, recent decisions, and active todos. |
| summarize-session | Returns a prompt guiding the agent to produce a structured session summary. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Active Sessions | Currently active agent sessions across all tools |
TDQS
Scored across 30 tools
Many tools have overlapping purposes, such as capture/extract/template for creating vault items, and project_context/generate_context/init/resume/status for context management. While descriptions clarify some boundaries, an agent may struggle to select the right tool for note creation or context retrieval, leading to misselection.
Most tool names use lowercase with underscores and follow a verb_noun pattern, but exceptions like skill_list_installed, graph_related, and env_facts break the convention. Simple verb names (read, write, search) coexist with compound names (snapshot_repo_state, graph_cross_project), creating a mixed but still readable naming scheme.
With 30 tools, the server exceeds the typical well-scoped range. Some tools could be consolidated (e.g., project context tools, capture/extract/template), and the sheer number increases cognitive load for an agent.
The tool surface covers core CRUD/lifecycle operations for vault items, tasks, skills, and contexts. Minor gaps exist, such as no explicit task deletion or project listing, but these can be worked around via search and prune.