Skip to content

fix: identify submitted server version in publish diagnostics #111

fix: identify submitted server version in publish diagnostics

fix: identify submitted server version in publish diagnostics #111

name: Detect invalid publish PRs
# Stage 1 of 2 (see close-invalid-publish-prs.yml for stage 2).
#
# Some contributors try to "publish" a server by opening a PR that adds files
# under servers/ or edits data/seed.json. That is not how publishing works -
# servers are published with the mcp-publisher CLI against the live registry
# API. This job only DETECTS such PRs. It runs from the (untrusted) PR context
# with a read-only token, so it cannot comment on or close anything. When it
# finds a match it records just the PR number as an artifact; the privileged
# stage-2 workflow picks that up via workflow_run and acts on it.
#
# We never check out or run the PR's code - we only read changed file paths.
on:
pull_request:
types: [opened, reopened]
permissions:
contents: read
pull-requests: read
jobs:
detect:
runs-on: ubuntu-latest
env:
GH_TOKEN: ${{ github.token }}
GH_REPO: ${{ github.repository }}
PR_NUMBER: ${{ github.event.pull_request.number }}
AUTHOR_ASSOCIATION: ${{ github.event.pull_request.author_association }}
steps:
- name: Classify PR
id: classify
run: |
set -euo pipefail
match=false
# Maintainers may legitimately touch these files (e.g. dev seed data),
# so never flag their PRs.
case "$AUTHOR_ASSOCIATION" in
MEMBER|OWNER|COLLABORATOR)
echo "Author association '$AUTHOR_ASSOCIATION' is trusted; not flagging."
;;
*)
# Use the paginated REST endpoint: `gh pr view --json files` caps
# at 100 files with no pagination, which could truncate a large PR
# and cause a false match.
mapfile -t FILES < <(gh api --paginate "repos/$GH_REPO/pulls/$PR_NUMBER/files" --jq '.[].filename')
if [ "${#FILES[@]}" -gt 0 ]; then
# Flag only if the PR touches *exclusively* publish-attempt files
# and nothing else, so that legit PRs touching seed data
# alongside real code are left alone.
#
# Keep this list in sync with the copy in stage 2. It stays an
# explicit list rather than a broad `**/server.json` glob:
# because we require *every* file to match, an over-broad
# pattern would auto-close a legitimate single-file PR.
# data/servers/** is here because #1524 put its server.json
# there - a natural guess next to data/seed.json - and slipped
# through the original servers/**-only list.
publish_attempt=0
other=0
for f in "${FILES[@]}"; do
if [[ "$f" == servers/* \
|| "$f" == data/servers/* \
|| "$f" == data/data/servers/* \
|| "$f" =~ ^data/[^/]*server\.json$ \
|| "$f" == "data/seed.json" \
|| "$f" == "server.json" \
|| "$f" == "servers.json" ]]; then
publish_attempt=1
else
echo "Non-publish file changed: $f"
other=1
fi
done
if [ "$publish_attempt" -eq 1 ] && [ "$other" -eq 0 ]; then
match=true
fi
fi
;;
esac
echo "PR #$PR_NUMBER match=$match"
echo "match=$match" >> "$GITHUB_OUTPUT"
if [ "$match" = "true" ]; then
echo "$PR_NUMBER" > pr-number.txt
fi
- name: Record flagged PR number
if: steps.classify.outputs.match == 'true'
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v4
with:
name: flagged-pr
path: pr-number.txt
retention-days: 1