Repository navigation
Commit a9c2fc7
fix: describe how the content is actually consumed
The overview claimed the content is consumed by "any MCP-capable client".
There is no MCP server in this repository and never has been: Cursor and
Claude Code read the files directly from a checkout. The line promised a
capability that does not exist, which is a defect rather than a wording
preference — a reader evaluating the project would go looking for a server
to register.
Describe the real mechanism instead: Cursor applies rules/*.mdc by scope
glob and takes skills by name; Claude Code reads skills/ and rules/ from the
workspace or a referenced checkout; any agent that can read workspace files
works the same way.
`pages.yml` also path-filters on a `mcp-tools.json` that has never existed in
this repository's history — it comes from the shared fleet site builder,
which supports tool repos that do ship an MCP server. Filed separately rather
than edited here.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Signed-off-by: TMHSDigital <154358121+TMHSDigital@users.noreply.github.com>1 parent 5a57d9d commit a9c2fc7
1 file changed
Lines changed: 1 addition & 1 deletion
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
39 | 39 | | |
40 | 40 | | |
41 | 41 | | |
42 | | - | |
| 42 | + | |
43 | 43 | | |
44 | 44 | | |
45 | 45 | | |
| |||
0 commit comments