skills-but-for-dev
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GITHUB_TOKEN | No | Optional GitHub personal access token for higher rate limits. If omitted, the server attempts to auto-discover a token via the GitHub CLI (`gh auth token`) or falls back to an unauthenticated, quota-conserving mode. |
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
} |
| logging | {} |
| prompts | {
"listChanged": true
} |
| resources | {
"subscribe": false,
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| pingA | Check if the skills-but-for-dev MCP server is operational and view version & token status. |
| search_github_reposA | Search for fresh, healthy, and license-compliant GitHub repositories for an architecture stack layer. When to use:
|
| get_repo_healthC | Fetch in-depth maintenance and responsiveness health metrics for a specific GitHub repository. |
| search_hf_modelsA | Search for open-source AI models on Hugging Face Hub with hardware/VRAM requirements. When to use:
|
| search_hf_spacesB | Search Hugging Face Spaces for running web demos, reference architectures, or prototypes. When to use:
|
| get_readmeA | Fetch and sanitize documentation for a GitHub repo or Hugging Face model/Space. IMPORTANT SAFETY: Output is sanitized against prompt injection and wrapped in <untrusted_content> tags. Giant benchmark tables and YAML frontmatter blocks are omitted to prioritize Usage and Quickstart sections. |
| compare_candidatesB | Compare a list of repository or model identifiers side-by-side in a structured comparison table. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 7 tools
Each tool maps to a distinct resource+action: repo health for a single repo, search across three separate sources (GitHub repos, HF models, HF spaces), readme fetching, comparison, and a server ping. The three search_* tools are cleanly partitioned by source type, and get_repo_health vs search_github_repos is clearly differentiated by single-repo inspection vs discovery.
Nearly all tools follow a verb_noun pattern (get_repo_health, search_github_repos, search_hf_models, search_hf_spaces, get_readme, compare_candidates). The bare 'ping' is a conventional outlier, and 'repo' vs 'github_repos' abbreviates the same concept differently across tools.
Seven tools is well-scoped for a discovery/evaluation server: three parallel search tools, two single-item fetchers, one comparison tool, and a health check. Every tool earns its place with no redundancy.
The discovery-to-evaluation workflow is well covered: search repos/models/spaces, read docs, check repo health, and compare candidates. Minor gap: no HF-side popularity/recency metrics tool (e.g. downloads/likes for models or spaces) equivalent to get_repo_health, though search results may partially compensate.