Machine-native repository of concluded operational experiences. One question. One file.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Fetch the body of a CorpusFetch record. Format 'md' (default, ~1500 tokens) or 'py' (zero-prose stdlib slice, ~300 tokens). Prefer warrant-first: call corpusfetch_fetch_warrant before this.
Slug of the record
Format: "md" (default, ~1500 tokens) or "py" (~300 tokens)
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
claude mcp add --transport http corpusfetch https://corpusfetch.com/mcp
Fetch the body of a CorpusFetch record. Format 'md' (default, ~1500 tokens) or 'py' (zero-prose stdlib slice, ~300 tokens). Prefer warrant-first: call corpusfetch_fetch_warrant before this.
| Argument | Type | Description |
|---|---|---|
| slug* | string | Slug of the record |
| format | string | null | Format: "md" (default, ~1500 tokens) or "py" (~300 tokens) |
{
"type": "object",
"title": "FetchRecordParams",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"required": [
"slug"
],
"properties": {
"slug": {
"type": "string",
"description": "Slug of the record"
},
"format": {
"type": [
"string",
"null"
],
"default": null,
"description": "Format: \"md\" (default, ~1500 tokens) or \"py\" (~300 tokens)"
}
}
}Fetch the ~220-token warrant (precondition seal) for a CorpusFetch record. Fetch this before the .md record. If the seal fails, do not fetch the record.
| Argument | Type | Description |
|---|---|---|
| slug* | string | Slug of the record (the URL path segment after /c/ and before .md) |
{
"type": "object",
"title": "FetchWarrantParams",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"required": [
"slug"
],
"properties": {
"slug": {
"type": "string",
"description": "Slug of the record (the URL path segment after /c/ and before .md)"
}
}
}Route a Python exception class to candidate record addresses in the CorpusFetch pillar. Returns warrant_url, py_url, md_url, and fetch_when for each candidate. Fetch the warrant first (~220 tokens); if its seal fails, do not fetch the .md.
| Argument | Type | Description |
|---|---|---|
| exception_class* | string | Python exception class, e.g. KeyError, TimeoutError, ValueError |
{
"type": "object",
"title": "LookupExceptionParams",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"required": [
"exception_class"
],
"properties": {
"exception_class": {
"type": "string",
"description": "Python exception class, e.g. KeyError, TimeoutError, ValueError"
}
}
}Route a runtime symptom to candidate record addresses in the CorpusFetch pillar. Returns warrant_url, py_url, md_url, and fetch_when for each candidate. Fetch the warrant first (~220 tokens); if its seal fails, do not fetch the .md.
| Argument | Type | Description |
|---|---|---|
| symptom* | string | Runtime symptom key, e.g. UNICODE_HOMOGLYPH_OR_MOJIBAKE, or a DOMAIN_* key |
{
"type": "object",
"title": "LookupSymptomParams",
"$schema": "https://json-schema.org/draft/2020-12/schema",
"required": [
"symptom"
],
"properties": {
"symptom": {
"type": "string",
"description": "Runtime symptom key, e.g. UNICODE_HOMOGLYPH_OR_MOJIBAKE, or a DOMAIN_* key"
}
}
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 2, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.