## Thinking Path
> - Paperclip manages AI agents that perform work.
> - Paperclip Runner connects durable task runs to local provider
processes.
> - The full-stack paid matrix exposed failures after the runner
integrity repair.
> - Verified JavaScript entrypoints lost their relative module graph
when Linux executed them through descriptor paths.
> - Returned provider startup errors also remained pending and became
indeterminate after recovery.
> - Sparse Codex tool lifecycle events lost the `write_document`
identity before task transcript projection.
> - This pull request repairs those three boundaries and makes the
structured-question fixture deterministic.
> - The benefit is repeatable provider startup, exact failure replay,
and correct inline Plan placement.
## Linked Issues or Issue Description
Refs #12721 and #12700.
**What happened?**
The paid runner matrix failed ACPX and OpenCode startup before provider
session creation. The runner journal then replaced the original startup
error with an indeterminate recovery result. Native Codex saved a Plan
but rendered it only as a fallback card. A legacy Claude waiting reply
could also echo the reserved terminal marker before the answer arrived.
**Expected behavior**
Verified JavaScript providers must start from immutable
descriptor-backed artifacts. Returned startup failures must persist as
terminal failed command results. Native tool lifecycle updates must
preserve the `write_document` boundary. Pre-answer fixture output must
not contain the reserved terminal marker.
**Steps to reproduce**
1. Run the local provider cells in the Runner Full-Stack E2E workflow.
2. Observe ACPX and OpenCode fail during `session.open` before provider
execution.
3. Observe recovery report `execution_indeterminate` instead of the
original startup error.
4. Run the native Codex Plan cell and observe the fallback Plan card
after the tool activity row.
5. Run the legacy Claude structured-question resume cell and observe an
early marker echo in waiting prose.
**Paperclip version or commit**
`0f9452101740835ce0b1488a204bf48acd5bafc3`
**Deployment mode**
Local development with the paid GitHub Actions acceptance workflow.
## What Changed
- Bundle the ACPX sidecar and OpenCode proxy as self-contained Node ESM
entrypoints before hashing and verified descriptor launch.
- Anchor ACPX dynamic provider package resolution at a
controller-derived provider-pack root and keep that root out of the
provider child environment.
- Persist executor-returned startup errors as redacted durable failed
command results while retaining indeterminate recovery for true process
death.
- Coalesce sparse native tool items by stable ID so a late
`write_document` name, input, and result reach the transcript boundary
once.
- Forbid the structured-question fixture from spelling or announcing its
reserved terminal marker before the user answers.
## Verification
- Rust and TypeScript regression tests cover durable failed replay, true
crash ambiguity, bundle closure, package-root derivation, environment
filtering, exact Codex tool lifecycle coalescing, and prompt
determinism.
- Local execution is intentionally limited to formatters and static diff
checks. GitHub Actions will run tests, type checks, builds, and security
checks.
- After ordinary CI is green, scoped paid cells will validate one ACPX
launch, one OpenCode launch, native Codex Plan projection, and legacy
Claude structured resume before a complete matrix rerun.
- Prior failing matrix:
https://github.com/paperclipai/paperclip/actions/runs/33682434315
## Risks
- Bundling changes the bytes covered by provider launch hashes.
Provider-pack generation already hashes the final built files.
- ACPX still loads qualified provider packages dynamically. The
controller supplies a normalized package root, while existing version,
digest, path, and descriptor checks remain active.
- Durable `failed` is terminal. Replays return the same redacted result
and do not execute the provider effect twice.
> 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 based on GPT-5 with agentic reasoning, repository
inspection, code editing, Git, parallel subagents, and GitHub Actions
coordination. The exact deployed snapshot and context-window size are
not exposed to this task.
## 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 linked related public work or described the bug in
this PR
- [x] I have not referenced internal or instance-local Paperclip issues
or links
- [x] My branch name describes the change and contains no internal
ticket id
- [ ] I have run tests locally and they pass (intentionally deferred to
GitHub Actions)
- [x] I have added or updated tests where applicable
- [x] No documentation change is required for this runtime repair
- [x] I have considered and documented the 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
## Thinking Path
> - Paperclip is the open source app people use to manage AI agents for
work
> - Agents are configured through `adapterConfig`, whose `env` block
holds the credentials an agent needs to talk to its provider (API keys,
tokens, and similar)
> - Those bindings come in several shapes: a legacy bare string, `{
type: "plain", value }`, and the indirection forms `{ type: "secret_ref"
}` / `{ type: "user_secret_ref" }`
> - Every endpoint that serializes an agent returned `adapterConfig` as
stored, so every `plain` binding was returned verbatim in the API
response
> - That means any caller able to read an agent — including the agent
itself via `GET /api/agents/me` — received live credentials in
plaintext, and those values then propagate into client state, logs, and
network traces
> - The exposure spans three response families that share no common
serializer: the single-agent detail reads, the company agent-list read,
and the create/update/lifecycle routes that echo the stored row straight
back
> - This pull request routes all three through one presenter that
redacts `plain` env bindings, so the leak is closed server-side and
cannot be bypassed by the caller
> - The benefit is that agent credentials stop appearing in API
responses while `secret_ref` indirection continues to work unchanged
## Linked Issues or Issue Description
No public upstream issue exists for this, so the underlying bug is
described inline below following
`.github/ISSUE_TEMPLATE/bug_report.yml`.
**What happened?**
Every endpoint that serializes an agent returned the full plaintext
value of each `adapterConfig.env` entry whose `type` was `"plain"` (and
each legacy bare-string binding). Any actor authorized to read an agent
received that agent's live credentials in the response body. Three
distinct response families were affected:
- **Single-agent reads** — `GET /api/agents/{id}` and `GET
/api/agents/me`, via `buildAgentDetail`.
- **Company agent list** — `GET /api/companies/{companyId}/agents`,
which serializes rows directly and therefore does not inherit any fix
applied to `buildAgentDetail`. Callers that pass the configuration-read
check received unredacted rows for every agent in the company in a
single request, making this the broadest of the three.
- **Mutation responses** — agent create, `PATCH /api/agents/{id}`, and
the `pause` / `resume` / `clear-error` / `approve` / `terminate` routes,
each of which echoes the stored row back to the caller.
**Expected behavior**
Read endpoints should never emit stored plaintext credentials. `plain`
bindings should be replaced with a redaction sentinel before
serialization, while `secret_ref` and `user_secret_ref` bindings — which
contain no secret material — pass through untouched.
**Steps to reproduce**
1. Configure an agent with an `adapterConfig.env` entry such as `{
"OPENAI_API_KEY": { "type": "plain", "value": "sk-example" } }`.
2. Call `GET /api/agents/{id}` (or authenticate as that agent and call
`GET /api/agents/me`).
3. Observe `sk-example` returned verbatim in the response body.
4. Call `GET /api/companies/{companyId}/agents` as a
configuration-reading caller and observe `sk-example` returned verbatim
for that agent alongside every other agent's credentials.
5. Call `PATCH /api/agents/{id}` with any unrelated field (for example
`{ "title": "Renamed" }`) and observe `sk-example` returned verbatim in
the mutation response.
**Paperclip version or commit**
Reproduced on `master` at `f12bb27b`.
**Deployment mode**
Self-hosted / local development server.
## What Changed
- `server/src/redaction.ts`: adds `redactAgentAdapterConfig`, which
rewrites every bare-string or `{ type: "plain", value }` env binding to
`{ type: "plain", value: "***REDACTED***" }` and passes `secret_ref` /
`user_secret_ref` bindings through unchanged. Reuses the existing
`REDACTED_EVENT_VALUE` and `isSecretRefBinding` /
`isUserSecretRefBinding` / `isPlainBinding` helpers — no new
dependencies.
- `server/src/redaction.ts`: `env` is destructured out and sanitized
only by `redactAgentEnvBinding`, while the remaining adapter keys go
through `redactEventPayload`. Previously the already-redacted `env` was
passed back through `sanitizeRecord`, so each binding was processed
twice — safe only because the sentinel is a fixed point of that second
pass. The two paths are now disjoint, making the invariant structural
rather than coincidental.
- `server/src/routes/agents.ts`: `buildAgentDetail` applies
`redactAgentAdapterConfig` before serialization, so `GET
/api/agents/{id}` and `GET /api/agents/me` both redact at the response
layer. Restricted views inherit the same protection.
- `server/src/routes/agents.ts`: adds `redactAgentRowForResponse`, the
single presenter for every response that emits a raw agent row, and
applies it to the company agent-list route and to the create / update /
pause / resume / clear-error / approve / terminate routes. It composes
with `redactForRestrictedAgentView` rather than replacing it: that
helper is an authorization filter (blank the whole config for low-trust
actors), this one is secret hygiene (mask values for every actor scope),
and the two invariants stay independent. `buildAgentDetail` now
delegates to the same presenter instead of inlining the call.
- `server/src/routes/agents.ts`: adds `restoreRedactedAgentEnv` on the
PATCH path so a client that round-trips a redacted detail response back
through `PATCH /api/agents/{id}` does not zero out stored values —
redacted-sentinel entries matching an existing key are restored from
storage.
## Verification
- `pnpm --filter @paperclipai/server exec tsc --noEmit` — clean.
- `pnpm --filter @paperclipai/server exec vitest run
agent-permissions-routes.test.ts` — 57 tests pass.
- Adjacent suites (`redaction`, `agent-adapter-validation-routes`,
`agent-cross-tenant-authz-routes`, `agents-pending-approval-config`,
`agents-service-secret-bindings`, `built-in-agent-routes`,
`plugin-managed-agents`, `agent-skills-routes`) — 8 files, 72 tests
pass, no regressions.
- Both new route tests were confirmed to **fail** with the route changes
reverted and pass with them applied, so they genuinely pin the behaviour
rather than passing incidentally.
Tests added:
- `server/src/__tests__/redaction.test.ts`: covers legacy-string, `{
type: "plain" }`, `secret_ref`, and `user_secret_ref` bindings,
asserting the plaintext value never appears in the serialized result;
plus coverage that non-env adapter keys are still sanitized, that env
binding shapes survive intact, and that configs with no `env` block are
handled.
- `server/src/__tests__/agent-permissions-routes.test.ts`: `GET
/api/agents/{id}` asserts redaction rather than plaintext passthrough;
new `GET /api/agents/me` redaction test across the same binding shapes;
new test asserting the `PATCH` round-trip preserves stored values; new
test asserting the board `GET /api/companies/{companyId}/agents`
response redacts every binding shape; new test asserting a mutation
response redacts rather than echoing the stored plaintext.
No real secret values appear in any test, fixture, or commit message.
## Risks
- **Behavioral change for API consumers.** Any client that read a
plaintext credential out of an agent detail, agent-list, or mutation
response will now receive `***REDACTED***`. This is the intended
security fix, but it is a breaking change for such consumers, which must
move to `secret_ref` indirection.
- **Mutation responses are redacted too.** Callers that previously
relied on a create or update response to echo back the credential they
had just written must now read it from their own request. This is
consistent with the `restoreRedactedAgentEnv` round-trip path, which
already assumes the client holds a redacted copy.
- **Round-trip data loss, mitigated.** A client that GETs an agent and
PATCHes the object straight back would otherwise persist the sentinel
over the real value. `restoreRedactedAgentEnv` restores redacted entries
from storage; the round-trip is covered by a regression test. A PATCH
that *intentionally* sets a value literally equal to the sentinel is not
distinguishable and would be treated as "unchanged" — an acceptable
trade-off given the sentinel is not a plausible credential.
- **No migration.** Stored data is untouched; redaction happens purely
at serialization time, so the change is fully reversible by revert.
- **Overlap with existing PRs** — see the duplicate-search note below.
Maintainers may prefer to consolidate rather than merge this in
isolation.
## Model Used
Claude Opus 4.8 (`claude-opus-4-8`), extended thinking enabled, with
tool use and local test execution.
## Duplicate search
Searching open and closed PRs for prior art surfaced several overlapping
efforts against the same defect. Linking them for maintainer triage — I
am not claiming this PR supersedes them, and consolidation may well be
preferable:
- #9823 — `fix(security): redact adapterConfig secrets on all agent read
endpoints` (closest overlap)
- #8779 — `fix(server): redact agent config secrets in read and mutation
responses`
- #8330 — `fix(server): redact adapterConfig.env for cross-actor agent
reads`
- #4856 — `fix(server): redact adapter env secrets in agent API
responses`
- #4763 — `fix(server): redact adapter_config secrets in agent detail
responses`
- #1839 — `fix: redact secret env vars from agent API responses`
- #4967 — `fix(routines): redact adapterConfig.env in GET
/api/routines/{id}` (same class, routines surface)
## 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 searched the GitHub PR list (open and closed) for similar or
duplicate 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)
- [ ] My branch name describes the change and contains no internal
Paperclip ticket id — **not met**: the branch and title carry an
internal ticket id. Renaming the branch would invalidate this PR; happy
to reopen from a clean branch if maintainers prefer.
- [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 (no
user-facing docs affected)
- [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 —
the one P2 (env entries processed twice) is addressed above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
Co-authored-by: Matthew Glover <5413384+glovario@users.noreply.github.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
Co-authored-by: Andrew Aymeloglu <aaymeloglu@gmail.com>
## Thinking Path
> - Paperclip is the control plane for agents that perform work.
> - Paperclip Runner connects durable provider sessions to individual
task runs through PRP.
> - Provider continuity and per-run authority are different lifetimes.
> - The existing implementation mixed those lifetimes and lost event
metadata between provider frames, runnerd, persistence, API
sanitization, and the task thread.
> - That caused failed continuation, missing progress and Plans,
duplicate replies, hidden failures, and unsafe recovery.
> - This repair gives every heartbeat fresh authority, preserves
qualified provider-session continuity, and restores one lossless
presentation path without changing direct adapters.
## Linked Issues or Issue Description
**What happened?**
A second native heartbeat could reuse tickets, leases, command receipts,
sequence state, and run identity from the first heartbeat. Provider
phase and item identity could be lost before the UI read them. Redaction
could corrupt protocol discriminators while still missing malformed
credential tails. The task thread could fold progress into the final
response, hide failures, or show more than one final answer. Native
Codex also exposed approval modes that do not yet have a durable
approval bridge.
**Expected behavior**
Each heartbeat uses a new PRP authority epoch. Codex and OpenCode
preserve exact qualified provider sessions; ACPX emits an explicit
continuity event when its qualified process-replacement policy is used.
Every accepted provider event is presented, classified as internal, or
surfaced as unsupported. The task page shows chronological progress,
reasoning summaries, activity, Plans, interactions, terminal failures,
and exactly one final reply. Direct adapters retain their existing path.
**Steps to reproduce**
1. Enable the unified experimental Paperclip Runner setting.
2. Create a local native Codex, OpenCode, ACPX Claude, or ACPX Codex
agent.
3. Run response, Plan, structured-question/resume, restart,
cancellation, and failure scenarios.
4. Reload the task while active, waiting, failed, and settled.
5. On the old implementation, observe stale run authority, missing
classifications, incomplete output, or duplicated/folded replies.
**Paperclip version or commit**
The repair is based directly on `master` at
`87d05e194b643810d16d20612115acd01d735d43`.
**Deployment mode**
Local development with the embedded database.
Related work: Refs #12616, #12646, #12666, #12685, and #12700.
## What Changed
- Rotates PRP control-plane, outbox, ticket, lease, command, receipt,
and sequence authority for each heartbeat while carrying forward only a
validated provider-session identity.
- Reads `control-plane-state.json`, validates both durable schemas and
lifecycle values, resumes coherent current runs, archives qualified
settled authority, and quarantines malformed or mismatched scoped state
without moving ambiguous live legacy state.
- Preserves Codex provider phase and stable item identities so
commentary remains progress and only `final_answer` becomes final.
- Adds raw OpenCode HTTP/SSE boundary coverage and canonical reasoning
lifecycle mapping.
- Makes ACPX normalization lossless for visible reasoning, tool
lifecycle metadata, stable bounded identities, Plan revisions,
structured requests, failures, and qualified process replacement. Only
the compatible terminal assistant message is promoted as final.
- Applies schema-aware redaction before generic JWT-shaped detection and
scans every diagnostic string leaf. Malformed raw/escaped quoted
credential tails are redacted in both server and durable Rust state.
- Restores snapshot-style chronological task presentation, expandable
tool activity, inline Plan cards, visible waiting/resume/cancel/failure
states, and exactly one final answer.
- Makes `never` the only qualified native Codex permission mode and
rejects unsupported persisted native modes with remediation. OpenCode
and ACPX policies remain intact.
- Keeps the unified experimental Runner setting as the only enablement
flag. Onboarding and direct Codex, Claude, and OpenCode stay on their
legacy execution/finalization paths.
- Adds cross-language goldens, authority/recovery/fault coverage, exact
response/count assertions, and native plus legacy acceptance scenarios.
## Verification
- Pull-request GitHub Actions run Rust formatting/tests, TypeScript
checks, server/UI tests, builds, protocol drift checks, browser E2E, and
security scans.
- A separate workflow-only validation ref is pinned directly on this PR
head and runs the 35-cell paid local matrix: three core scenarios plus
structured-question resume and restart/resume for native Codex, native
OpenCode, ACPX Claude, ACPX Codex, and direct Codex/Claude/OpenCode.
Run: https://github.com/paperclipai/paperclip/actions/runs/33682434315
- Acceptance requires exact single visible replies, monotonic sequences,
matching envelope discriminators, one semantic terminal, one run
terminal, no unresolved interaction, no duplicate mutation, no secret
leakage, provider continuity, and zero native rows for direct adapters.
- Per maintainer direction, tests are running in GitHub Actions rather
than on the slower local host. Only formatters and static diff checks
were run locally.
## Risks
- Recovery from old or partial filesystem state is sensitive. The repair
fails closed, preserves active or unverifiable authority, and
quarantines only state whose scoped ownership is safe to move.
- Provider event formats can change. Closed validators and boundary
goldens turn new or malformed events into visible diagnostics instead of
silent drops.
- Shared task presentation could affect direct adapters. Runtime-fact
gating plus the direct-adapter matrix protect the existing path.
- Managed and remote providers are not qualified here. Shared code
continues to compile and fail safely, but live qualification is
deferred.
> 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 based on GPT-5. The exact deployed snapshot and
context-window size are not exposed to this task. It used agentic
reasoning, repository inspection, code editing, Git, parallel subagents,
and GitHub Actions.
## 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 and contains no internal
Paperclip ticket id
- [ ] I have run tests locally and they pass (intentionally deferred to
GitHub Actions)
- [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 risks above
- [ ] All Paperclip CI gates are green
- [ ] The paid local-provider matrix is 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
## Thinking Path
> - Paperclip is the open source app people use to manage AI agents for
work.
> - The runner package is useful only when the application can start,
observe, and recover a native Codex run safely.
> - Existing direct adapters must keep their current execution and
finalization paths.
> - The application boundary therefore needs additive persistence,
authorization, coordination, and recovery behind an explicit
experimental adapter.
> - This pull request adds that Codex-only boundary without activating
generalized providers, remote environments, or the later task/SDK
surfaces.
## Linked Issues or Issue Description
**Subsystem affected**
Shared contracts, database persistence, adapter utilities, server
native-runtime services, and the experimental Paperclip Runner adapter.
**Problem or motivation**
The already-landed runner package has a qualified Codex path, but the
application needs durable native-run state, guarded runtime selection,
authenticated coordination, tool security, finalization, and recovery
before the experimental adapter can be exercised safely.
**Proposed solution**
Add a Codex-only `paperclip_runner` application path behind the existing
default-off native-runner setting. Bind native state and coordination to
company/run identity, preserve persisted-run recovery, and leave every
direct adapter on its existing legacy execution path.
**Alternatives considered**
The earlier stack boundary introduced a generalized executor and
remote-environment lifecycle here. That made this PR depend on
implementations in higher PRs and changed reusable sandbox behavior
globally. Those pieces are now deferred together to #12592.
**Roadmap alignment**
ROADMAP.md does not list a conflicting native-runner integration
project. This change adds the application boundary for the existing
Runner architecture.
## What Changed
- Added native run/result/finalization/provider-trace persistence,
shared validators, and idempotent migration/replay coverage.
- Added guarded Codex-only runtime selection, authenticated PRP
coordination, recovery, finalization, and interaction services.
- Added run/company-bound tool-gateway authorization, credential
redaction, SSRF protections, and replay-safe behavior.
- Added the explicit `paperclip_runner` adapter behind the default-off
rollout setting.
- Preserved legacy answered-question wake projection and direct-adapter
execution/finalization paths.
- Hardened cancellation so only owned in-memory child processes are
signaled; persisted recycled PIDs/process groups are never trusted.
- Retained the narrow Claude ACPX isolated-context security follow-up
discovered after #12590.
- Deferred the generalized executor, provider ingress, remote lifecycle,
SDK/lab/eval work, release-process changes, and lockfile.
## Verification
- Changed-file delta against `master`: 133 files.
- GitHub Actions is the authoritative verification environment for this
PR.
- Full CI, security, and Greptile review will run on this lowest
unmerged stack PR.
- Local tests/build/typecheck were not run because this checkout is
resource constrained.
- Static diff/reference checks pass, and `pnpm-lock.yaml` is unchanged.
## Risks
- This touches central heartbeat and agent-route code, so legacy
compatibility is the primary risk.
- Runtime selection remains Codex-only and explicit; direct Codex,
Claude, OpenCode, process, HTTP, and plugin adapters remain on their
existing paths.
- Fresh native starts fail closed while the rollout flag is off;
persisted native records remain readable and recoverable.
- Cancellation, company/run binding, tool calls, status decisions, and
completion writes are guarded or replay-safe.
> 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.
## Model Used
OpenAI Codex, GPT-5.6, with repository tools, code execution, and
parallel agent review.
## 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 linked existing issues or described the issue in-PR
following the relevant issue template
- [x] I have not referenced internal or instance-local Paperclip issues
or links
- [x] My branch name describes the change and contains no internal
Paperclip ticket id
- [ ] I have run tests locally and they pass — GitHub Actions is
authoritative for this resource-constrained checkout
- [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 risks above
- [ ] All Paperclip CI and security gates are green
- [ ] Greptile is 5/5 with no open actionable findings
- [x] I will address all Greptile and reviewer comments before merge
## Stack
- Position: 3 of 5 overall; lowest of 3 currently unmerged
- Base: `master`
- Previous:
[#12590](https://github.com/paperclipai/paperclip/pull/12590), qualified
Claude ACPX runtime — merged
- Next: [#12592](https://github.com/paperclipai/paperclip/pull/12592),
generalized Codex executor, task experience, and developer SDKs
---------
Co-authored-by: Dev Agent <dev@paperclip.ing>
## Thinking Path
> - Paperclip is the open source app people use to manage AI agents for
work.
> - Apps give those agents governed access to external tools.
> - Remote MCP setup needs secure endpoint validation and durable
credentials.
> - PostHog needs both browser sign-in and personal API key setup paths.
> - This pull request adds the shared remote MCP foundation and the
PostHog definition.
> - The benefit is a secure and reusable base for later app connection
work.
## Linked Issues or Issue Description
Refs #11965
This is stack 1 of 11. It replaces the first reviewable part of #11965.
## What Changed
- Add guarded remote MCP setup and credential handling.
- Add PostHog OAuth and API key connection methods.
- Add focused server, shared contract, and UI coverage.
- Keep the migration replay-safe and idempotent.
- Give the late-close security regression the same 10-second CI headroom
as the adjacent real-timer handshake test.
- Synchronize fake-timer handshake tests at the exact ensure-session
boundary so real filesystem setup cannot race the fake deadline.
- Drive PTY overflow coverage only after listener registration so
scheduling cannot reorder the test fixture.
## Verification
- pnpm exec vitest run
packages/adapter-utils/src/acpx-engine/execute.test.ts
server/src/__tests__/plugin-worker-manager.test.ts (220 passed; affected
cases also passed five focused stress repetitions)
- `pnpm exec vitest run
packages/adapter-utils/src/acpx-engine/execute.test.ts -t "never leaks a
sandbox-provided value from a late close rejection into logs or the
result"` (1 passed)
- `pnpm exec vitest run
packages/adapter-utils/src/acpx-engine/execute.test.ts -t "never
promotes a late ensureSession resolution|closes a late-resolving real
handle exactly once"` (2 passed)
- `pnpm -r typecheck`
- `pnpm --filter @paperclipai/server exec vitest run
src/__tests__/tool-access-service.test.ts`
- `pnpm --filter @paperclipai/db check:migrations`
- `pnpm build`
## Risks
- Remote endpoint validation can reject configurations that previously
passed without checks.
- OAuth configuration errors can block setup until the operator corrects
the provider settings.
- The migration uses guarded statements so repeated execution is safe.
- The test-only synchronization changes do not affect runtime behavior;
they remove filesystem/fake-clock and listener-registration races
observed under parallel CI load.
> I checked `ROADMAP.md`. This stack continues the existing app
connection work from #11965 and does not duplicate another planned item.
## Model Used
OpenAI Codex, GPT-5. The runtime model ID and context window were not
exposed. The model used reasoning, tool use, 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 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>
## Thinking Path
> - Paperclip is the control plane that coordinates autonomous agent
work.
> - Agents need to collaborate on issues beyond their current
assignment.
> - Cross-issue comments and updates are useful, but an unbounded run
can create cascading side effects.
> - The control plane must preserve company-wide collaboration while
containing each run's influence.
> - Comment attribution must also show the responsible user and the
acting agent in audits.
> - This pull request adds run-bound cross-issue containment,
attribution, and agent-class wake rules.
> - The benefit is safer collaboration without restoring issue-assignee
ownership restrictions.
## Linked Issues or Issue Description
**What existing behavior does this improve?**
Agent-authenticated issue comments, updates, reopen behavior, and
assignee wake routing.
**Subsystem affected**
Cross-cutting: server routes and services, shared contracts, database
schema and migration, and implementation documentation.
**Current behavior**
An authenticated agent can collaborate across company issues, but one
heartbeat run has no per-run side-effect boundary. Comment records also
do not persist the responsible user separately from the acting agent.
**Proposed behavior**
Require a valid heartbeat run for agent cross-issue comments and
updates. Audit each attempt and cap a run at 20 cross-issue effects.
Keep the cap in log-only mode until it automatically changes to
enforcement at 2026-08-11 00:00 UTC. Preserve same-issue writes. Use
agent-class wakes for agent comments. Keep same-run completion comments
from reopening completed work. Record the responsible user on
agent-authored comments and activity.
**Reason and benefit**
Agents can collaborate on other issues without an assignment gate, while
each run has an atomic and inspectable side-effect limit. Operators can
identify both the acting agent and the responsible user.
**Breaking changes**
After 2026-08-11 00:00 UTC, the twenty-first cross-issue comment or
update from one heartbeat run returns a containment error. Agent
cross-issue writes without valid run context are rejected. The migration
is additive and backfills existing agent-authored comment attribution
where the source data is available.
## What Changed
- Added an atomic per-run counter for cross-issue agent comments and
updates.
- Added audit events for allowed and rejected cross-issue effects.
- Added the automatic log-only to enforcement flip at 2026-08-11 00:00
UTC.
- Added responsible-user attribution to agent-authored comments,
activity records, shared types, and validators.
- Added an additive migration and migration coverage for existing
comments.
- Updated reopen, resume, and wake behavior so agent comments create
agent-class wakes and same-run completion comments remain inert.
- Updated the implementation specification and regression coverage.
## Verification
- `pnpm exec vitest run
server/src/__tests__/cross-issue-influence-limit.test.ts
server/src/__tests__/issue-comment-attribution-audit-routes.test.ts
server/src/__tests__/issue-comment-reopen-routes.test.ts
packages/db/src/issue-comment-on-behalf-migration.test.ts` — 97 tests
passed.
- `pnpm -r typecheck` — passed, including migration safety checks.
- `pnpm test:run` — server batch: 3,364 passed and 2 skipped; UI batch:
3,504 passed. One unrelated CLI doctor test warned because this agent
runtime injects static AWS credentials.
- `env -u AWS_ACCESS_KEY_ID -u AWS_SECRET_ACCESS_KEY pnpm exec vitest
run cli/src/__tests__/secrets.test.ts` — 8 tests passed and confirmed
the CLI failure was ambient-environment sensitive.
- `pnpm build` — passed.
## Risks
- The fixed enforcement timestamp changes production behavior
automatically on 2026-08-11 00:00 UTC. Audit logs before that time
provide rollout visibility.
- The per-run counter serializes on the heartbeat-run row. This prevents
concurrent attempts from racing past the cap but adds a small lock scope
for cross-issue writes.
- Existing comments can only be backfilled when their acting run or
agent attribution is recoverable.
> 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 GPT-5 in the Codex agent runtime. The runtime did not expose a
context-window size. Reasoning, shell tools, code editing, and test
execution were enabled.
## 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>
## Thinking Path
> - Paperclip is the control plane for autonomous AI companies, so agent
work needs visible ownership, recovery, and operator controls.
> - This local branch had accumulated several related control-plane
reliability and operator-experience fixes across recovery actions,
watchdog folding, model-profile defaults, mentions, markdown editing,
plugin launchers, and small UI polish.
> - The branch needed to be converted into a PR against the current
`origin/master` without losing dirty work or including lockfile/workflow
churn.
> - The safest standalone shape is a single rollup PR because the
recovery/server/UI files overlap heavily across the local commits and
splitting would create avoidable conflicts.
> - This pull request replays the local branch onto latest
`origin/master`, preserves the uncommitted work as logical commits, and
adds a Zod 4 validator compatibility fix found during verification.
> - The benefit is that the May 17 local branch can be reviewed and
merged as one coherent, conflict-free branch under the 100-file Greptile
limit.
## What Changed
- Rebased the local May 17 branch work onto current `origin/master` in a
dedicated worktree.
- Preserved and committed previously dirty changes for recovery retry
handling, plugin/sidebar launcher polish, and `.herenow` ignores.
- Added recovery-action behavior for returning source issues to `todo`
when retrying source-scoped recovery.
- Included the existing local recovery/liveness/watchdog fold, Codex
cheap-profile, markdown/mention, duplicate-agent, and UI polish commits
from the branch.
- Normalized shared validator `z.record(...)` schemas to explicit
string-key records for Zod 4 compatibility.
- Confirmed the PR has no `pnpm-lock.yaml` or `.github/workflows/*`
changes and stays below the 100-file Greptile limit.
## Verification
- `pnpm install --frozen-lockfile --ignore-scripts`
- `npm run install` in
`node_modules/.pnpm/sqlite3@5.1.7/node_modules/sqlite3` to build the
local native sqlite3 binding after installing with scripts disabled
- `pnpm exec vitest run packages/shared/src/validators/issue.test.ts
packages/shared/src/project-mentions.test.ts
packages/adapter-utils/src/server-utils.test.ts
server/src/__tests__/heartbeat-model-profile.test.ts
server/src/__tests__/issue-recovery-actions.test.ts
server/src/__tests__/issue-agent-mutation-ownership-routes.test.ts
server/src/__tests__/heartbeat-active-run-output-watchdog.test.ts
server/src/__tests__/plugin-local-folders.test.ts
ui/src/components/IssueRecoveryActionCard.test.tsx
ui/src/components/Sidebar.test.tsx
ui/src/components/SidebarAccountMenu.test.tsx
ui/src/components/IssueProperties.test.tsx
ui/src/components/MarkdownEditor.test.tsx
ui/src/components/MarkdownBody.test.tsx
ui/src/lib/duplicate-agent-payload.test.ts
ui/src/pages/Routines.test.tsx`
- First pass: 13 files passed with 201 passing tests; 3 server files
failed before sqlite3 native binding was built.
- After rebuilding sqlite3:
`server/src/__tests__/heartbeat-model-profile.test.ts`,
`server/src/__tests__/issue-recovery-actions.test.ts`, and
`server/src/__tests__/heartbeat-active-run-output-watchdog.test.ts`
passed/loaded; embedded Postgres tests were skipped by the local host
guard.
- `pnpm --filter @paperclipai/shared typecheck`
- `pnpm --filter @paperclipai/adapter-utils typecheck`
- `pnpm --filter @paperclipai/server typecheck`
- `pnpm --filter @paperclipai/ui typecheck`
## Risks
- Medium risk: this is a broad rollup PR across recovery semantics,
server tests, shared validators, and UI surfaces.
- Some embedded Postgres tests skipped locally due the host guard, so CI
should provide the stronger database-backed signal.
- UI changes were covered by component tests, but no browser screenshot
was captured in this PR creation pass.
- This branch may overlap with existing recovery/liveness PR work; merge
this PR independently or restack/close overlapping branches rather than
merging duplicate implementations together.
> 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-based coding agent, tool-enabled local repository
and GitHub workflow, medium reasoning effort.
## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [ ] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
## Thinking Path
> - Paperclip orchestrates AI-agent companies through a control plane
that can start, supervise, and recover agent runs.
> - Local adapters are the bridge between Paperclip issues and concrete
agent runtimes such as Claude, Codex, and other ACP-compatible tools.
> - The roadmap calls out broader “bring your own agent” and claw-style
agent support, and ACPX gives Paperclip one path to normalize multiple
ACP agents behind a single adapter.
> - The branch needed to become one reviewable PR against current
`paperclipai/paperclip:master`, without carrying stale base conflicts or
generated lockfile churn.
> - This pull request adds an experimental built-in `acpx_local`
adapter, integrates it through the server/CLI/UI adapter surfaces, and
adds regression coverage for runtime execution, skill sync, stream
parsing, diagnostics, and log redaction.
> - The benefit is that Paperclip can run Claude/Codex/custom ACP agents
through ACPX while keeping operator configuration, skills, logging, and
transcript rendering inside the existing adapter model.
## What Changed
- Added `@paperclipai/adapter-acpx-local` with server execution, config
schema, ACPX session handling, CLI formatting, UI config helpers, and
stdout parsing.
- Registered `acpx_local` across CLI, server, shared constants, UI
adapter metadata, adapter capabilities, and agent creation/editing
surfaces.
- Added ACPX runtime execution support with persistent sessions,
local-agent JWT environment handling, skill snapshots, runtime skill
materialization, and isolation/security regressions.
- Added ACPX adapter diagnostics and marked the adapter experimental in
the UI.
- Added command/env secret redaction for resolved command metadata in
adapter-utils, server event storage, and the Agent Detail invocation UI.
- Added Storybook coverage for ACPX config, transcript rendering, and
skill states, plus PR screenshots under `docs/pr-screenshots/pap-2944/`.
- Rebased the branch onto current `public-gh/master`; `pnpm-lock.yaml`
is intentionally not included and there are no migration/schema changes.
## Verification
- `pnpm exec vitest run
packages/adapters/acpx-local/src/server/execute.test.ts
packages/adapters/acpx-local/src/server/test.test.ts
packages/adapters/acpx-local/src/cli/format-event.test.ts
packages/adapters/acpx-local/src/ui/parse-stdout.test.ts
packages/adapter-utils/src/server-utils.test.ts
server/src/__tests__/redaction.test.ts
server/src/__tests__/acpx-local-execute.test.ts
server/src/__tests__/acpx-local-skill-sync.test.ts
server/src/__tests__/acpx-local-adapter-environment.test.ts
server/src/__tests__/adapter-routes.test.ts
server/src/__tests__/agent-skills-routes.test.ts
ui/src/adapters/metadata.test.ts` — 12 files, 87 tests passed.
- `pnpm --filter @paperclipai/adapter-acpx-local typecheck` — passed.
- `pnpm --filter @paperclipai/server typecheck` — passed.
- `pnpm --filter @paperclipai/ui typecheck` — passed.
- Confirmed PR diff does not include `pnpm-lock.yaml`, database schema
files, or migrations.
Screenshots:
## Risks
- Medium risk: this introduces a new built-in adapter package and
touches runtime execution, adapter registration, agent config, skills,
and transcript rendering.
- ACPX and ACP agent behavior can vary by installed tool versions; the
adapter is marked experimental to set operator expectations.
- `pnpm-lock.yaml` is excluded per repository PR policy, so dependency
lock refresh must be handled by the repo’s automation or maintainers.
- No database migration risk: no schema or migration files changed.
> 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 coding agent based on GPT-5, with repository tool use,
shell execution, git operations, and local verification. Exact hosted
context window was not exposed in this environment.
## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge
---------
Co-authored-by: Paperclip <noreply@paperclip.ing>
Add AES-256-GCM local encrypted secrets provider with auto-generated
master key, stub providers for AWS/GCP/Vault, and a secrets service
that normalizes adapter configs (converting sensitive inline values to
secret refs in strict mode) and resolves secret refs back to plain
values at runtime. Extract redaction utilities from agent routes into
shared module. Redact sensitive values in activity logs, config
revisions, and approval payloads. Block rollback of revisions
containing redacted secrets. Filter hidden issues from list queries.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Dotta
Paperclip DevOps Engineer
Forgotten
