gapfind
Read-only audit of which backend capabilities (REST routes, MCP tools) a repo's webapp never surfaces, plus webapp calls to routes that do not exist. Trigger when the user says "gapfind" followed by a repo name, or asks "are there api endpoints or mcp tools in the backend that are not surfaced / used in webapp pages", or what the backend can do that the UI does not show. Report and propose only, never auto-fix. Do NOT trigger for mock data / dead buttons (that is fakefind), fleet-standard compliance (drift-scan / assfix), or coverage against the wrapped software's API (wrappee-drift).
- License 许可
- MIT
- 最近更新
- 2026-11-14
- Stars
- ★ 2
- Forks
- 0
01 前言 Front Matter
- name
- gapfind
- description
- Read-only audit of which backend capabilities (REST routes, MCP tools) a repo's webapp never surfaces, plus webapp calls to routes that do not exist. Trigger when the user says "gapfind" followed by a repo name, or asks "are there api endpoints or mcp tools in the backend that are not surfaced / used in webapp pages", or what the backend can do that the UI does not show. Report and propose only, never auto-fix. Do NOT trigger for mock data / dead buttons (that is fakefind), fleet-standard compliance (drift-scan / assfix), or coverage against the wrapped software's API (wrappee-drift).
03 SKILL.md 正文
gapfind
Find backend services the webapp forgot. Most fleet repos were built MCP-server-first and got a webapp later, so useful endpoints and tools often have no page.
Canonical procedure
Read first and follow its phases, status meanings and classification categories:
D:\Dev\repos\mcp-central-docs\patterns\GAPFIND_AUDIT_SOP.md
Run the script, do not reason from memory:
uv run --project D:\Dev\repos\<repo> python D:\Dev\repos\mcp-central-docs\scripts\gapfind.py D:\Dev\repos\<repo> --out D:\Dev\repos\<repo>\reports\gapfind-<yyyy-mm-dd>.md
Date from the real clock. Check the detection notes at the top of the output before trusting anything below them.
Hard rules
Read-only. Report and propose only. Fixes are a separate, approved pass.
Code truth, not a skim. Route and tool lists come from the live objects. Never answer the question by sampling a few files.
0 routes detected = detection failure, not a clean result. Rerun with
--app module:attr / --syspath DIR.
Classify every non-used row as SURFACE / HEADLESS / ADMIN / DEAD / DUPLICATE /
REVERSE (definitions in the SOP). The script's status column is a heuristic; via-REST?
twins in particular must be confirmed by reading the route handler.
Output
The script report, then a classification table, then a prioritized fix list: existing
endpoints that only need UI first (cheapest user value), new backend work last. Suggest
.gapfind-headless entries for everything classified HEADLESS so reruns stay clean.
04 同仓库更多 Skills
Docs Research
Use MCP Central Docs server to search, ask, and retrieve documentation (FastMCP, standards, packaging).
wrappee-drift
Read-only, single-repo audit of a repo's tool coverage against the software it wraps (its current real API/docs, not training-data memory) and against comparable/competing wrappers. Trigger when the user says "wrappee drift" followed by a repo name, "capability audit" followed by a repo name, or asks whether a repo is missing features the wrapped software or a competitor now has. Always scoped to one named repo per invocation -- never trigger this fleet-wide or unprompted; that is explicitly against the SOP. Do NOT trigger for internal fleet-standard compliance (that is drift-scan) or strategic worth/archive decisions (that is quality-check) -- this is specifically about the gap between the repo's tool list and the wrapped software's/competitors' actual current capabilities.
modules
暂无简介
mcp-standards
Use when developing MCP servers to access comprehensive standards for scaffolding, packaging, sampling, documentation, testing, error handling, monitoring, and CI/CD workflows.
webapp-cua-test
Run the pre-Tauri CUA click-through test (real browser, dev stack, OCR "Connected" wait, sidebar nav walk) for a fleet MCP server repo under D:\Dev\repos. Trigger when the user says "cua webapp test" or "webapp clickthrough" followed by a repo name, asks for an e2e click-through of a webapp, or asks whether a webapp boots and connects in a real browser. Not for the installed NSIS app (that is the nsis-build skill / cua-nsis-test) and not for DOM assertions (Playwright), and not for hunting fake data or dead buttons (fakefind).
ship
Final release gate for a fleet MCP server repo under D:\Dev\repos -- run all quality gates, build the release artifact, verify it, commit, push, and cut a GitHub release. Trigger this whenever the user says "ship it", "ship" followed by a repo name, "release" followed by a repo name, or asks to finalize and publish work that's already been done in a repo. This is the terminal step after assfix/nsis-build, not a substitute for them -- if the repo hasn't been assessed recently, say so before shipping rather than shipping blind.