skill-forge-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SKILL_FORGE_PERSIST | No | Set to 'true' to persist progress to ~/.skill-forge-mcp/state.json |
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 |
|---|---|
| search_processA | Search all phases of the skill creation process by keyword. Case-insensitive partial match. Returns matching phase IDs, section names, and matched lines. |
| mark_progressB | Record progress status for a phase. status: 'not-started' | 'in-progress' | 'completed'. |
| get_statusA | Return a summary of all phase progress (not-started/in-progress/completed) and access counts. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| create_skill | Guide the agent through the full skill creation process (Phase 0→8). Provide a topic and the agent will follow the structured workflow. |
| resume_skill | Resume a skill creation session. Checks current progress and continues from where you left off. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| manifest | List of all 9 phases. Overview, dependencies, and section structure for each phase. Read this first to understand the overall picture. |
| phase-0 | SKILL.md structure, frontmatter, Progressive Disclosure, invocation control. First time only. |
| phase-1 | Check existing skills, mechanism selection (CLAUDE.md / Prompt File / Skill), baseline measurement. |
| phase-2 | Define 'good', research authoritative theories, organize agent information needs, design workflow. |
| phase-3 | Verify Phase 2 deliverables against 6 axes: theory→practice, completeness, judgment, verifiability, technology specificity, concreteness. |
| phase-4 | Fill Phase 3 gaps. Patterns/code examples, standards/specs, techniques, anti-patterns, verification, tech stack specifics, concrete examples. |
| phase-5 | Organize research results and verify all 8 categories are covered: theory, info requirements, process, patterns, anti-patterns, verification, tech stack, examples. |
| phase-6 | Condense all research into a practical skill of 500 lines or fewer. Filter → design decisions → distillation → writing. |
| phase-7 | Place the skill in the appropriate location and verify specification compliance. Placement, cross-surface, security, validation. |
| phase-8 | Verify the skill actually works and improve it. Evaluation scenarios, A/B testing, iteration loop, troubleshooting. |
TDQS
Scored across 3 tools
Each tool has a clearly distinct role: get_status reads overall progress, search_process searches content across phases, and mark_progress updates a phase status. There is no meaningful overlap or ambiguity between them.
All tool names follow a consistent imperative verb_noun pattern: get_status, search_process, mark_progress. The naming style is uniform, clear, and predictable.
Three tools is on the smaller end for a full server, but it is a reasonable scope for a focused progress-tracking workflow. Each tool serves a necessary function without redundancy.
The core workflow is covered: check progress, search phase content, and update status. A minor gap is the lack of a direct full-phase retrieval tool, but the available tools cover the main use case without dead ends.