fbc-guard
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| FBC_GUARD_ALLOW_PRIVATE | No | Set to 1 to allow checking pages on your own machine or intranet (localhost and private-network URLs). By default, fetching is limited to public hosts. |
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
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| check_fbc_codeA | Statically check HTML or JavaScript that reads |
| check_fbc_urlA | Fetch a landing page and run the same checks as check_fbc_code on its HTML plus every same-origin it loads, so tracking code kept in an external file is included. Cross-origin scripts (tag managers, the Meta Pixel, analytics SDKs) are never fetched and are listed under skipped_scripts. Limits: 30 s timeout per request, at most 20 scripts, 2 MB per file. Uses a mobile Safari user agent. Only public hosts are fetched: URLs that resolve to loopback, private, link-local or other internal addresses are refused, on every redirect hop too (start the server with FBC_GUARD_ALLOW_PRIVATE=1 to check local pages). Returns JSON with |
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 2 tools
The two tools are clearly separated by input mode: check_fbc_code analyzes supplied source text, while check_fbc_url fetches a live page and its same-origin scripts. Sharing the same rule set does not create ambiguity because the descriptions make the execution boundary explicit.
Both names follow the same predictable pattern: check_fbc_ followed by the input type (code or url). The snake_case convention is consistent across the entire server.
Two tools is below the typical 3–15 range, but the server has a narrow, single-purpose domain and each tool covers a genuinely distinct workflow. The surface is minimal without being redundant, though it is still on the thin side.
The server covers both obvious entry points for this domain: pasted source code and deployed URL scanning, including same-origin external scripts. It returns detailed per-rule results, evidence, warnings, and statuses for pass/fail/unknown/not_applicable, with no obvious core checker operation missing.