Files
kz-tob 43d7af8064 gh-cli: intercept GitHub fetches from MCP fetch tools (#274)
* gh-cli: intercept GitHub fetches from MCP fetch tools

The fetch hook only matched WebFetch, so a GitHub URL fetched through an
MCP fetch tool bypassed it entirely. Match `mcp__.*[Ff]etch` as well, which
covers Exa's `web_fetch_exa` and equivalents from other MCP servers.

A matcher alone is not enough: WebFetch passes a single `url`, while MCP
fetch tools pass a `urls` array and batch several pages into one call. Read
both shapes. A tool call is atomic, so one GitHub URL anywhere in a batch
denies the whole call, and each offending URL is labeled with its own
suggestion. Single-URL calls keep the message they had.

Split the URL classification into suggest_api, suggest_raw, and
suggest_github_com behind a suggest_for_url dispatcher so it can run per URL
in a loop, and collapse eight verbatim copies of the clone hint into one
helper. Behavior is unchanged; blob and tree merge into one branch because
they emitted identical text.

The plugin README's interception table now also lists the pull, issues,
releases, and gist patterns the hook already handled but never documented.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

* gh-cli: address review on the MCP fetch interceptor

Test the matcher. Nothing exercised the regex that decides whether the
hook runs at all, so a matcher firing on nothing would still pass every
other suite here. matcher.bats reads it out of hooks.json and checks it
against real tool names, and refuses to run against an empty matcher,
which would make every match assertion pass vacuously.

Widen the matcher to fetch, scrape, crawl, and extract. Firecrawl's
scrape and Tavily's extract retrieve a URL like any fetch tool but carry
no "fetch" in the name, so they bypassed the hook while the README
promised "any MCP fetch tool". The README now names what matches.

Check a string-valued `urls`. A server declaring `urls: string | string[]`
sent a bare string, `arrays` dropped it, and the fetch went out
unauthenticated. Listing the field twice keeps it under both shapes.
`prompt` is still not scanned, so a prompt mentioning a GitHub URL does
not false-deny.

Give the closing note its own line when several URLs are denied; it
previously trailed only the last entry.

Guard assert_suggestion_starts_with against an empty reason and an empty
prefix, either of which let it pass while inspecting nothing.

Document the api.github.com contents, releases, and actions rows. The
generic `gh api` row was actively wrong for /contents/, pointing readers
at the anti-pattern the shim exists to block.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-25 10:38:13 -04:00

134 lines
4.4 KiB
Bash

#!/usr/bin/env bats
# Tests for the PreToolUse matchers in hooks.json.
#
# The hook scripts are well covered, but nothing exercised the matcher that
# decides whether they run at all — so a matcher that fires on nothing would
# still pass every other suite in this directory. These tests read the matcher
# out of hooks.json and check it against real tool names.
#
# Claude Code evaluates a matcher containing regex metacharacters as an
# unanchored JavaScript regular expression. `grep -E` is likewise unanchored,
# and these patterns use only constructs common to both.
setup() {
HOOKS_JSON="${BATS_TEST_DIRNAME}/hooks.json"
FETCH_MATCHER=$(jq -r '.hooks.PreToolUse[]
| select(.hooks[].command | contains("intercept-github-fetch.sh"))
| .matcher' "$HOOKS_JSON")
BASH_MATCHER=$(jq -r '.hooks.PreToolUse[]
| select(.hooks[].command | contains("intercept-github-curl.sh"))
| .matcher' "$HOOKS_JSON")
}
# A matcher that failed to parse would be empty, and an empty pattern matches
# every tool name — which would make every assertion below pass vacuously.
assert_matcher_nonempty() {
if [[ -z "$1" ]]; then
echo "Matcher not found in $HOOKS_JSON — every match assertion would pass vacuously"
return 1
fi
}
assert_matches() {
local pattern="$1" tool="$2"
assert_matcher_nonempty "$pattern" || return 1
if ! printf '%s\n' "$tool" | grep -Eq "$pattern"; then
echo "Expected matcher '$pattern' to match tool '$tool'"
return 1
fi
}
refute_matches() {
local pattern="$1" tool="$2"
assert_matcher_nonempty "$pattern" || return 1
if printf '%s\n' "$tool" | grep -Eq "$pattern"; then
echo "Expected matcher '$pattern' NOT to match tool '$tool'"
return 1
fi
}
# =============================================================================
# The matchers are present at all
# =============================================================================
@test "matcher: hooks.json is valid JSON" {
run jq -e . "$HOOKS_JSON"
[[ $status -eq 0 ]]
}
@test "matcher: fetch hook has a matcher" {
assert_matcher_nonempty "$FETCH_MATCHER"
}
@test "matcher: bash hook has a matcher" {
assert_matcher_nonempty "$BASH_MATCHER"
}
# =============================================================================
# Built-in tools
# =============================================================================
@test "matcher: fetch matcher matches WebFetch" {
assert_matches "$FETCH_MATCHER" "WebFetch"
}
@test "matcher: bash matcher matches Bash" {
assert_matches "$BASH_MATCHER" "Bash"
}
@test "matcher: fetch matcher does not match Bash" {
refute_matches "$FETCH_MATCHER" "Bash"
}
@test "matcher: fetch matcher does not match Read" {
refute_matches "$FETCH_MATCHER" "Read"
}
@test "matcher: fetch matcher does not match WebSearch" {
refute_matches "$FETCH_MATCHER" "WebSearch"
}
# =============================================================================
# MCP tools that retrieve a URL — these are the ones the hook exists for
# =============================================================================
@test "matcher: fetch matcher matches Exa web_fetch_exa" {
assert_matches "$FETCH_MATCHER" "mcp__exa__web_fetch_exa"
}
@test "matcher: fetch matcher matches a bare fetch server" {
assert_matches "$FETCH_MATCHER" "mcp__fetch__fetch"
}
@test "matcher: fetch matcher matches Firecrawl scrape" {
assert_matches "$FETCH_MATCHER" "mcp__firecrawl__firecrawl_scrape"
}
@test "matcher: fetch matcher matches Firecrawl batch scrape" {
assert_matches "$FETCH_MATCHER" "mcp__firecrawl__firecrawl_batch_scrape"
}
@test "matcher: fetch matcher matches a crawl tool" {
assert_matches "$FETCH_MATCHER" "mcp__exa__crawling_exa"
}
@test "matcher: fetch matcher matches Tavily extract" {
assert_matches "$FETCH_MATCHER" "mcp__tavily__tavily_extract"
}
# =============================================================================
# MCP tools that do not retrieve a URL — no reason to run the hook
# =============================================================================
@test "matcher: fetch matcher does not match Exa search" {
refute_matches "$FETCH_MATCHER" "mcp__exa__web_search_exa"
}
@test "matcher: fetch matcher does not match context7 docs query" {
refute_matches "$FETCH_MATCHER" "mcp__context7__query-docs"
}
@test "matcher: fetch matcher does not match a Slack reader" {
refute_matches "$FETCH_MATCHER" "mcp__slack__slack_read_channel"
}