DottaandPaperclip
7a8df30953
## Thinking Path
> - Paperclip lets people manage AI agents for work.
> - Paperclip Runner must preserve each agent's model and thinking
effort.
> - The Grok migration in #17922 passes explicit effort to native
Runner.
> - The pinned Grok 1.0.13 CLI reports native session metadata instead
of standard ACP config options.
> - The standard effort guard rejects that valid session before the
first prompt.
> - This pull request binds model and effort on Grok session open and
verifies the native reply.
> - Restored sessions keep the selected effort instead of reverting to
Grok's default.
## Linked Issues or Issue Description
Related: https://github.com/paperclipai/paperclip/pull/17922
**What happened?**
A migrated Grok 4.7 agent with High effort fails at native session.open
before it can run a tool. Grok 1.0.13 has no standard configOptions
response and rejects session/set_config_option. Its native session
metadata supports the requested model and effort. A cold load also
resets effort to High unless the request supplies it again. The upgrade
banner also calls Grok “Codex”.
**Expected behavior**
The native session must acknowledge the selected model and effort before
a prompt. A cold restore must preserve both values. The upgrade banner
must describe the selected harness accurately.
**Steps to reproduce**
1. Upgrade a Grok agent to Paperclip Runner with explicit High effort.
2. Assign a bounded shell task. The pre-fix native session fails during
admission.
3. For the restore case, create a Low native session, close the process,
and load it without effort metadata. The pinned CLI reports High.
**Paperclip version or commit**
Observed at f60ac29cf1 with Grok 1.0.13.
**Deployment mode**
Cloud, with the board operator interface.
## What Changed
- Project the bound model and effort into native Grok session/new,
session/load, and session/resume metadata before transport writes and
observation.
- Require a correlated native acknowledgement with the exact model,
effort, and session identity. Reject conflicting metadata, selection
drift, stale connections, and prompts before admission.
- Connect lazy restored handles through read-only native session
inspection. Keep the existing process ownership and launch snapshot
checks.
- Replace Grok's unsupported standard config setter. Keep the existing
Claude effort path.
- Add behavioral guard cases and a real ACPX client fixture for fresh
and cold Low/High sessions. The fixture resets effort on restore if
metadata is absent, as the pinned CLI does.
- Publish Copilot profile v69 because its source identity includes the
shared runtime adapter. Preserve published v66/v67/v68 bytes and the
reviewed Grok v67 candidate as history. Keep old profiles closed for
launch.
- Explain the native protocol in the Grok Runner guide and use
provider-independent upgrade banner text.
## Verification
- Three native create/load/resume cases failed before the repair. The
pre-fix guard rejected the native metadata reply.
- Seven affected Runner test files: 234 tests passed after the master
merge. These include native Grok admission, the real ACPX transport
fixture, runtime ownership, Claude effort, and profile history.
- The review follow-up extends the same Low/High fixture through a
complete first turn after cold restore. It checks the repeated native
load, original session id, actual response event, and completed turn.
All 127 tests in the three affected Grok/adapter files passed again;
Runner typecheck and source identity checks also passed.
- All 167 existing agent form tests passed. Token gates and diff checks
passed.
- Workspace typecheck passed. Final Runner typecheck and generated
profile checks passed.
- Production `pnpm build` passed, including after the master conflict
resolution. Runner typecheck, profile history checks, and the 234
affected tests passed on the merged source. All 55 Rust ACPX library
tests also passed on the merged source.
- Real pinned Grok 1.0.13 metadata probes reproduced the unsupported
standard setter and native create/load/resume behavior. No model prompt
was sent. An earlier actual ACPX runtime probe passed fresh and cold
Low/High metadata checks. A repeat with an explicitly isolated HOME
failed authentication before session creation, so it does not count as
final live acceptance. The deterministic transport fixture is green; a
deployed shell task remains required.
- `pnpm test:run` did not complete locally. Embedded Postgres failed
during bootstrap after five attempts. This is the same local fixture
limitation recorded for #17922. Fresh PR CI must pass before merge.
- After deployment, retry the original bounded shell task with the saved
Grok 4.7/High configuration. Verify actual tool output, a successful
run, and Done state after reload.
- Final commit `810d84fd19728e7d798989582f26b05f69166db2`: all CI gates
passed. Greptile and the formal Paperclip review both reached 5/5. No
unresolved review threads remain.
## Risks
- This depends on the pinned Grok 1.0.13 native metadata format. Missing
or conflicting replies stop admission instead of silently changing
effort.
- Cold restores must make one read-only native inspection call before
admission. The existing handshake timeout and process ownership checks
still apply.
- The shared adapter changes Copilot's source identity. The new current
profile includes the final source hashes; the prior profile remains
decodable but cannot launch.
- This does not change the Grok CLI version, account selection, search
policy, dependency patch, or lockfile.
## Model Used
OpenAI Codex, GPT-6. The exact serving model ID and context window are
not exposed in this session. The agent used reasoning, code editing,
shell tools, and browser tools.
## 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
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
10 KiB
Grok Build native runner
Select Grok Build in the native runner provider selector. The stored contract is
adapterType: "paperclip_runner" with provider: "acpx", acpxAgent: "grok",
and model: "grok-4.7". Existing grok_local agents keep their legacy adapter
until an explicit upgrade. On a qualified macOS ARM64 or Linux x64 computer,
Harness / Runtime → Upgrade runner preserves the agent, model, account,
instructions, and history. It starts fresh execution sessions. Advanced also
offers an explicit Runner choice. Unsupported legacy settings are rejected
before saving; select compatible settings or keep Legacy.
New Grok runner agents default to Full auto (approve all)
(acpxPermissionMode: "approve-all") in setup and the configuration form.
API configurations that omit the permission mode use the same default. No
additional permission setting is needed for unattended execution. Explicitly
saved restrictions remain unchanged.
Grok Build's qualified harness does not require the enableNativeRunner flag.
Explicit thinking effort uses reasoningEffort (minimal, low, medium,
high, or xhigh). The immutable v10 execution input, provider descriptor,
and recovery identity bind this value as grokEffort. The Grok provider boundary
adds native _meta.modelId and _meta.reasoningEffort to session create, load,
and resume requests before transport writes and observation. Grok 1.0.13 reports
the selected effort in _meta["x.ai/sessionConfig"].options; it does not support
the standard session/set_config_option effort API. The Runner checks that native
acknowledgement and the effective model before a prompt. A lazy restored handle
must connect through read-only native session inspection first. Reapplying the
bound metadata on restore prevents Grok from resetting effort to its default.
Missing configuration, a rejected value, or effort drift stops admission.
Omitting effort retains Grok's provider default. This setting remains outside
the model-visible task prompt. Historical execution inputs retain their schemas.
The upgrade resolves an unset model or the legacy grok-build default to the
native grok-4.7 default. Task-level model and effort overrides use the Grok
catalog and keep the account and tool policy unchanged.
Migration also preserves the effective disableWebSearch setting, whose legacy
default is true. Native input v10 carries it as grokDisableWebSearch through
startup and recovery. Grok's private requirements.toml enforces disabled search
above compatible project settings. Explicit false retains enabled search;
unset native configurations retain their provider default. See Grok's
configuration reference.
Grok Build speaks ACP over stdio.
The runner owns grok agent --no-leader stdio through ACPX, including session
identity, cancellation, recovery and the authenticated Paperclip MCP bridge.
ACP permission requests are approved by the runner under the default full-auto
policy. It does not add --always-approve: permission decisions remain under
the selected ACPX policy. Grok's ACP metadata cannot independently establish
Paperclip tool authority, so explicitly selecting approve-paperclip or
approve-reads returns the approval-required outcome; deny-all rejects requests.
Full auto does not bypass Paperclip's company permissions, governed approvals,
or execution-environment boundaries. Isolated ask
rules override project allow rules, and compatible always-approve settings are
locked off. Compatible hook/MCP discovery and shell login capture are disabled.
Installation and identity
Grok support ships inside the native runner and the public Paperclip server npm
artifact. There is no separate Grok npm package, binary payload, or npm lifecycle
download. The built-in launcher is identified as builtin:grok-acp version 1;
its native runtime identity is native:grok version 1.0.13. The historical
agentServerPackage/agentRuntimePackage wire fields carry these identities,
not npm dependencies. Existing npm-backed ACP bridges retain their package pins.
Provision Grok Build 1.0.13 at
/opt/paperclip/providers/grok/1.0.13/grok in the selected execution environment.
The sandbox provisioning helper is explicit and is never run by npm:
sudo node packages/paperclip-runner/scripts/provision-grok.mjs /opt/paperclip/providers/grok/1.0.13/grok
The standard Daytona image provisions it separately from the provider pack.
Custom images and local execution hosts must provide the same prerequisite.
The runner verifies the native executable checksum before credential refresh or
ACP startup; a missing prerequisite reports the required path and version.
Only macOS arm64 and Linux x64 are qualified. No ambient grok from PATH is used.
ACP must report the requested exact model; mismatches fail closed.
The distribution identity change deliberately rejects resume bindings from the former private-package profile. Start a fresh session after upgrading that unreleased profile; do not silently reinterpret its saved identity.
Instructions use Grok ACP session rules. Assigned skills live in the isolated Grok home. Steering and goals are unsupported. Token and cost values remain unknown when Grok does not report them; missing usage is not zero usage.
Authentication
Use the existing Grok company connection/login flow for subscription execution.
An explicitly selected company-secret XAI_API_KEY selects paid API execution.
There is no automatic subscription-to-API fallback. Remote execution cannot
borrow the operator's home credentials. Local execution can use the operator's
existing Grok login when no company login has been selected.
Only the selected credential is staged in the private runtime home. An ownership
lease fences concurrent processes. Before ACP startup, an expiring subscription
credential is refreshed through the verified Grok executable’s non-inference
models command. This bounded step suppresses output and prevents Grok 1.0.13
from caching a pre-refresh model list. Exact model verification still precedes
any prompt; refresh failure requires reconnecting Grok Build. After the provider exits, refreshed credentials
are copied back through Grok's existing identity and refresh checks. Runtime
credentials, refresh handoffs, and diagnostic logs are excluded from workspace
backups and removed on close. Session history remains available for resume. Host
configuration, other provider credentials and unselected keys are not forwarded
to Grok.
Evaluation
The private paperclip-evals repository maintains live-acpx-grok-4.7.json and
rosters/live-acpx-grok.json. The roster covers all 39 protocol cases. Its campaign
lane remains disabled pending complete live qualification.
Product E2E exposes runner-acpx-grok in core local/Daytona compatibility and
local session integrity. The explicit grok-qualification suite covers replies,
planning approvals, structured questions, downloadable project revisions, stop/resume
and continuation after controller restart
in both environments. Run with --suite grok-qualification; it is excluded from
scheduled --all. Use the canonical Product E2E dashboard and Evalbook reports.
The separate grok-subscription-qualification suite covers the same workflows
with the explicit GROK_AUTH_JSON fixture credential. It seeds only the disposable
company's private login home and supplies no API key. It does not exercise the
interactive sign-in UI. Authentication mode remains part of the profile identity.
After a subscription upgrade, a fresh grok login may be needed if the existing
login still reports the previous entitlement through ACP.
packages/paperclip-runner/scripts/grok-native-smoke.mjs records explicit auth,
model/binary identity, MCP outcome, durable resume and restrictive permissions.
Pass --auth subscription --auth-file /private/path/auth.json --output /private/report.json
or --auth api --output /private/report.json with an explicitly supplied key.
Each attempt is retained; unknown usage and cost are null.
Qualification requires the full protocol roster, selected product workflows, subscription and API execution locally and on Daytona, and three successful repetitions of core tool, approval and resume cases. Deterministic tests or a single successful browser task do not establish that qualification.
Remote verification
Do not run Docker on a developer laptop when using remote verification. The
Docker Runner check workflow offers maintainer-authorized manual Linux image
builds and broad source checks without provider credentials. It records the
source revision, resolved lock digest and immutable image reference. Paid
Product E2E remains behind the protected default-branch workflow and environment.
The public npm consumer check uses a digest-pinned, unprivileged container with
no checkout or credentials mounted. It downloads dependencies with lifecycle
scripts disabled and freezes the resulting consumer lockfile. It completes the
clean install with offline npm rebuild, running the deferred lifecycle hooks
without re-resolving bundled optional dependencies. Networking stays disabled,
the lockfile must remain unchanged, and a sentinel proves scripts actually ran.
The pinned image includes native build tools and local Node headers so dependency
hooks can compile without network access. Both executable admission probes run
in the same isolation. The verification user provisions Grok inside the disposable
test directory without privilege elevation or host /opt changes. Only the
positive probe mounts that binary read-only at the canonical sandbox path.
The Cloud application image also carries the controller-owned provider pack and
sets PAPERCLIP_RUNNER_REMOTE_PROVIDER_PACK_PATH. Remote ACPX execution verifies
the sandbox against that pack before using it, or stages the matching pack when
needed. The Cloud controller image does not install the native Grok executable;
the selected sandbox image must provide the prerequisite above.
Cloud builds must supply the full source SHA through PAPERCLIP_BUILD_COMMIT
to produce that verified pack. Unstamped local Cloud builds still work for other
features, but omit the pack and cannot start remote ACPX sessions.