ci(runner): allow trusted branch targets (#12768)

## Thinking Path

> - Paperclip uses paid runner tests to qualify agent execution.
> - The runner workflow controls provider secrets and AWS runner access.
> - The trusted workflow must stay on the protected default branch.
> - The code under test often exists on a branch before merge.
> - CODEOWNERS need a safe way to select that branch.
> - This pull request separates workflow authority from the code under
test.
> - The benefit is pre-merge AWS testing without target-controlled
workflow code.

## Linked Issues or Issue Description

**What existing behavior does this improve?**

The manual Runner Full-Stack E2E workflow can test only the default
branch.

**Subsystem affected**

GitHub Actions and the paid runner E2E security boundary.

**Current behavior**

A CODEOWNER must merge runner changes before the trusted AWS workflow
can test them.
Selecting another branch as the workflow ref is rejected.

**Proposed behavior**

A CODEOWNER starts the workflow from `master` and supplies a
same-repository branch in `target_branch`.
The authorization job resolves the branch to one commit SHA.
Catalog, image, and paid test jobs check out that SHA after
authorization.
Report sanitization and AWS publication use the trusted workflow SHA.

**Reason and benefit**

This permits paid pre-merge qualification on AWS.
It keeps the workflow definition, report sanitizer, history publisher,
environment deployment, and runner-group permission on `master`.

**Breaking changes**

None.
The new input is optional.
An omitted input still tests the default branch.

## What Changed

- Add the optional `target_branch` workflow input.
- Resolve only a branch in `paperclipai/paperclip` to an immutable SHA.
- Pin catalog, image, paid test, and Daytona provenance to the target
SHA.
- Pin report sanitization and AWS history publication to the trusted
workflow SHA.
- Disable persisted checkout credentials in every job.
- Key cancellation by the selected target branch.
- Add policy regression coverage and operator documentation.

## Verification

- `pnpm test:e2e:runner:unit` passes with 65 tests.
- `actionlint -ignore SC2129
.github/workflows/runner-full-stack-e2e.yml` passes.
- Prettier checks pass for all changed files.
- `git diff --check` passes.

## Risks

A CODEOWNER can authorize selected branch code to receive a cell-scoped
provider credential.
This is the intended trust decision.
The workflow rejects fork refs and target-controlled workflow
definitions.
The trusted workflow SHA owns report sanitization and AWS history
publication.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

OpenAI Codex, GPT-5.
The exact serving snapshot and context-window size are not exposed.
The model used tool-enabled reasoning and code execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge
This commit is contained in:
Dotta 2026-09-03 10:38:56 -05:00 committed by GitHub
parent 6b625425a5
commit 98c569b2df
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
4 changed files with 146 additions and 28 deletions

View File

@ -5,6 +5,10 @@ on:
- cron: "47 8 * * 0"
workflow_dispatch:
inputs:
target_branch:
description: "Branch in paperclipai/paperclip to test; the trusted workflow still runs from master"
type: string
required: false
all:
description: "Run the complete paid matrix when no narrower selector is supplied"
type: boolean
@ -38,10 +42,10 @@ permissions:
contents: read
concurrency:
group: runner-full-stack-e2e-${{ github.ref }}
# Development-only validation refs supersede older runs on the same ref.
# Preserve every protected default-branch campaign for its paid audit trail.
cancel-in-progress: ${{ github.ref != format('refs/heads/{0}', github.event.repository.default_branch) }}
group: runner-full-stack-e2e-${{ inputs.target_branch || github.event.repository.default_branch }}
# Development branch campaigns supersede older runs for the same target.
# Preserve every default-branch campaign for its paid audit trail.
cancel-in-progress: ${{ github.event_name == 'workflow_dispatch' && inputs.target_branch != '' && inputs.target_branch != github.event.repository.default_branch }}
jobs:
authorize:
@ -55,6 +59,7 @@ jobs:
test_runner: ${{ steps.runner.outputs.runner }}
max_parallel_default: ${{ steps.runner.outputs.max_parallel_default }}
max_parallel_limit: ${{ steps.runner.outputs.max_parallel_limit }}
target_sha: ${{ steps.target.outputs.sha }}
steps:
- name: Require default branch and allowlisted numeric actor IDs
env:
@ -89,6 +94,27 @@ jobs:
fi
done
- name: Resolve requested repository branch to an immutable commit
id: target
env:
GH_TOKEN: ${{ github.token }}
REPOSITORY: ${{ github.repository }}
TARGET_BRANCH: ${{ inputs.target_branch || github.event.repository.default_branch }}
run: |
set -euo pipefail
if [ -z "$TARGET_BRANCH" ] || [[ "$TARGET_BRANCH" == refs/* ]]; then
echo "target_branch must name a branch in this repository without a refs/ prefix." >&2
exit 1
fi
encoded_branch="$(jq -rn --arg branch "$TARGET_BRANCH" '$branch | @uri')"
target_sha="$(gh api -X GET "repos/$REPOSITORY/branches/$encoded_branch" --jq .commit.sha)"
if ! [[ "$target_sha" =~ ^[0-9a-f]{40}$ ]]; then
echo "The requested repository branch did not resolve to a commit." >&2
exit 1
fi
echo "sha=$target_sha" >> "$GITHUB_OUTPUT"
echo "Resolved the requested repository branch to $target_sha."
- name: Select paid test runner
id: runner
env:
@ -130,6 +156,9 @@ jobs:
daytona_image_content_id: ${{ steps.daytona_image_content.outputs.content_id }}
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
with:
ref: ${{ needs.authorize.outputs.target_sha }}
persist-credentials: false
- uses: pnpm/action-setup@0977fd99725f1db4007ccb2928dbb4e90d06cc86 # v6
with:
@ -233,6 +262,9 @@ jobs:
content_id: ${{ steps.image.outputs.content_id }}
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
with:
ref: ${{ needs.authorize.outputs.target_sha }}
persist-credentials: false
- name: No Daytona image needed
id: local_only
@ -261,6 +293,7 @@ jobs:
NEEDS_DAYTONA: ${{ needs.catalog.outputs.needs_daytona }}
IMAGE_CONTENT_ID: ${{ needs.catalog.outputs.daytona_image_content_id }}
IMAGE_TAG: ghcr.io/paperclipai/paperclip-daytona-runner:e2e-content-${{ needs.catalog.outputs.daytona_image_content_id }}
TARGET_SHA: ${{ needs.authorize.outputs.target_sha }}
run: |
set -euo pipefail
if [ "$NEEDS_DAYTONA" != true ]; then
@ -277,7 +310,7 @@ jobs:
docker buildx build \
--platform linux/amd64 \
--build-arg "PAPERCLIP_RUNNER_CONTENT_ID=${IMAGE_CONTENT_ID}" \
--build-arg "PAPERCLIP_RUNNER_SOURCE_REVISION=${GITHUB_SHA}" \
--build-arg "PAPERCLIP_RUNNER_SOURCE_REVISION=${TARGET_SHA}" \
--file docker/daytona-runner/Dockerfile \
--tag "$IMAGE_TAG" \
--push \
@ -354,6 +387,7 @@ jobs:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
with:
ref: ${{ needs.authorize.outputs.target_sha }}
persist-credentials: false
- uses: pnpm/action-setup@0977fd99725f1db4007ccb2928dbb4e90d06cc86 # v6
@ -416,7 +450,7 @@ jobs:
report:
name: Merge and enforce campaign result
if: always() && needs.catalog.result == 'success'
needs: [catalog, daytona_image, test]
needs: [authorize, catalog, daytona_image, test]
outputs:
history_source_ready: ${{ steps.history_source_ready.outputs.ready }}
runs-on: ubuntu-latest
@ -425,6 +459,10 @@ jobs:
contents: read
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
with:
# Reporting and sanitization are part of the trusted workflow boundary.
ref: ${{ github.sha }}
persist-credentials: false
- uses: pnpm/action-setup@0977fd99725f1db4007ccb2928dbb4e90d06cc86 # v6
with:
@ -511,7 +549,7 @@ jobs:
publish_history:
name: Publish pruned immutable history and landing site
needs: [catalog, report]
needs: [authorize, catalog, report]
if: always() && needs.catalog.result == 'success' && needs.report.outputs.history_source_ready == 'true'
runs-on: ubuntu-latest
timeout-minutes: 15
@ -525,6 +563,10 @@ jobs:
name: runner-e2e-history
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
with:
# Never execute target-controlled publication code with AWS credentials.
ref: ${{ github.sha }}
persist-credentials: false
- uses: pnpm/action-setup@0977fd99725f1db4007ccb2928dbb4e90d06cc86 # v6
with:

View File

@ -266,13 +266,35 @@ auto-stop/archive/delete values remain as cancellation backstops.
## GitHub Actions
`Runner Full-Stack E2E` has only `schedule` and `workflow_dispatch` triggers; it
never runs for a pull request or ordinary push. Because this repository is
public, manual campaigns fail before checkout unless they run from the default
branch and both the original actor and rerun actor have numeric GitHub user IDs
in the non-empty JSON-array repository variable
`RUNNER_E2E_ALLOWED_ACTOR_IDS`. Usernames are intentionally not trusted.
The first scheduled attempt is trusted automation; any human rerun of a
scheduled campaign must pass the triggering-actor allowlist.
never runs for a pull request or ordinary push. Start the trusted workflow from
the default branch. A CODEOWNER can set the optional `target_branch` input to
any branch in `paperclipai/paperclip`. The authorization job resolves that
branch to one immutable commit before any checkout. Catalog, image, and paid
test jobs check out that exact commit. Report sanitization and AWS history
publication explicitly check out the trusted workflow commit. The workflow
definition, runner-group permission, and protected-environment deployment still
come from the default branch. Do not select the target branch in GitHub's **Use
workflow from** control.
Because this repository is public, manual campaigns fail before checkout unless
the trusted workflow runs from the default branch and both the original actor
and rerun actor have numeric GitHub user IDs in the non-empty JSON-array
repository variable `RUNNER_E2E_ALLOWED_ACTOR_IDS`. Keep this stable-ID list in
sync with the owners of `.github/**` in `.github/CODEOWNERS`. Usernames are
intentionally not trusted. The first scheduled attempt is trusted automation;
any human rerun of a scheduled campaign must pass the triggering-actor
allowlist.
For example, this command runs one branch cell through the trusted default-branch
workflow:
```bash
gh workflow run runner-full-stack-e2e.yml \
--ref master \
-f target_branch=fix/example \
-f all=false \
-f id=core-compatibility.runner-codex.local.message-marker
```
Create a protected `runner-e2e-paid` GitHub environment, restrict it to the
default branch, limit environment administration to trusted maintainers, and
@ -302,11 +324,14 @@ it, and require a fresh ephemeral instance for each job so one paid cell cannot
leave state for the next. Provider secrets remain protected by the stable-ID
authorization checks and the default-branch-only `runner-e2e-paid` environment;
the fleet itself is not an authorization boundary. These external fleet controls
are as important as the workflow checks in a public repository.
are as important as the workflow checks in a public repository. A CODEOWNER
dispatch is an explicit authorization to execute the selected repository branch
with the cell's scoped provider credential.
Non-default validation runs share a concurrency key per ref and cancel an older
run when a replacement is dispatched. Protected default-branch paid campaigns
are retained and are never auto-cancelled, preserving their audit trail.
Development branch campaigns share a concurrency key per target branch and
cancel an older run when a replacement is dispatched. Default-branch target
campaigns are retained and are never auto-cancelled, preserving their audit
trail.
GitHub Actions artifacts are access-controlled 30-day operational copies, not
the permanent public history. They retain packaged PNG/WebM and generated

View File

@ -9,15 +9,23 @@ changes.
## GitHub authorization
Set `RUNNER_E2E_ALLOWED_ACTOR_IDS` to a non-empty JSON array of numeric GitHub
user IDs, for example `[123456,789012]`. Resolve each ID from the authenticated
CLI and verify the login before adding it:
user IDs. Keep the list equal to the owners of `.github/**` in
`.github/CODEOWNERS`. For example, use `[123456,789012]`. Resolve each ID from
the authenticated CLI and verify the login before adding it:
```bash
gh api users/LOGIN --jq '{login,id}'
```
The paid workflows reject manual dispatches outside the default branch before
checkout. They verify both the original actor and triggering actor for every
The paid workflows reject manual dispatches when the workflow definition does
not come from the default branch. A trusted dispatcher may name any branch in
`paperclipai/paperclip` as the code under test. The authorization job resolves
that branch through the GitHub API and passes only its immutable commit SHA to
the catalog, image, and paid test checkouts. Report sanitization and AWS history
publication explicitly use the trusted workflow commit. Never run the workflow
definition from the target branch.
The workflows verify both the original actor and triggering actor for every
scheduled or manual attempt, including human reruns. Every
secret-bearing job repeats this check as its first step so GitHub's partial-job
rerun feature cannot bypass a successful predecessor authorization job. The
@ -78,14 +86,23 @@ fresh ephemeral instance for every job, prohibit persistent runner reuse, and
disable interactive SSH/debug access unless a separate incident procedure
explicitly authorizes it.
Changing the runner does not widen secret access. The paid workflow still has
only schedule and manual triggers, requires the protected default branch and
allowlisted stable actor IDs before checkout, repeats that authorization as the
first matrix step, and receives provider credentials only from the protected
Changing the runner does not widen who can authorize secret access. The paid
workflow still has only schedule and manual triggers, requires its trusted
definition to come from the protected default branch, requires allowlisted
stable actor IDs before checkout, and repeats that authorization as the first
matrix step. Provider credentials come only from the protected
`runner-e2e-paid` environment. The fleet selector is an exact workflow literal;
the only repository-controlled input is its boolean rollout switch, so
the only repository-controlled routing input is its boolean rollout switch, so
configuration cannot redirect a secret-bearing job to an arbitrary runner.
The optional target branch is code, not workflow authority. A CODEOWNER who
dispatches a target branch explicitly authorizes that branch's selected test
process to receive the cell's scoped provider credential. The workflow resolves
the target only inside the same repository, pins one SHA for the campaign, and
checks it out only after authorization. Target-controlled code cannot replace
the report sanitizer or the AWS history publisher. Fork refs and
target-controlled workflow definitions do not enter this path.
## AWS OIDC and S3
The AWS role trust policy should accept only GitHub's OIDC audience and the

View File

@ -75,6 +75,13 @@ describe("public repository paid workflow security", () => {
expect(authorizeJob).toContain(
"AWS_PAID_RUNNER_ENABLED: ${{ vars.RUNNER_E2E_AWS_ENABLED }}",
);
expect(authorizeJob).toContain(
"Resolve requested repository branch to an immutable commit",
);
expect(authorizeJob).toContain(
"repos/$REPOSITORY/branches/$encoded_branch",
);
expect(authorizeJob).toContain('echo "sha=$target_sha"');
expect(paidJob).toContain(
"runs-on: ${{ needs.authorize.outputs.test_runner }}",
);
@ -89,7 +96,34 @@ describe("public repository paid workflow security", () => {
'[ "$MAX_PARALLEL" -gt "$MAX_PARALLEL_LIMIT" ]',
);
expect(fullStack).toContain(
"cancel-in-progress: ${{ github.ref != format('refs/heads/{0}', github.event.repository.default_branch) }}",
"group: runner-full-stack-e2e-${{ inputs.target_branch || github.event.repository.default_branch }}",
);
expect(fullStack).toContain(
"cancel-in-progress: ${{ github.event_name == 'workflow_dispatch' && inputs.target_branch != '' && inputs.target_branch != github.event.repository.default_branch }}",
);
expect(
fullStack.match(
/ref: \$\{\{ needs\.authorize\.outputs\.target_sha \}\}/g,
),
).toHaveLength(3);
expect(fullStack.match(/ref: \$\{\{ github\.sha \}\}/g)).toHaveLength(2);
expect(fullStack.match(/persist-credentials: false/g)).toHaveLength(5);
expect(fullStack).not.toContain("ref: ${{ inputs.target_branch }}");
expect(fullStack).toContain(
"PAPERCLIP_RUNNER_SOURCE_REVISION=${TARGET_SHA}",
);
const reportJob = fullStack.slice(
fullStack.indexOf(" report:"),
fullStack.indexOf(" publish_history:"),
);
const historyJob = fullStack.slice(fullStack.indexOf(" publish_history:"));
expect(reportJob).toContain("ref: ${{ github.sha }}");
expect(reportJob).not.toContain(
"ref: ${{ needs.authorize.outputs.target_sha }}",
);
expect(historyJob).toContain("ref: ${{ github.sha }}");
expect(historyJob).not.toContain(
"ref: ${{ needs.authorize.outputs.target_sha }}",
);
for (const [secret, condition] of Object.entries({
OPENAI_API_KEY: "matrix.credentialName == 'OPENAI_API_KEY'",