Record the two classic document repair cells and preserve the distinction between original native grades and analytical regrading.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Accept the requested marker plus concrete blocker explanation, calibrate contradictory and future-condition replies, and clarify future fixture punctuation. Preserve every original paid failure and label retained-DOM replay separately from live qualification.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
## Thinking Path
> - Paperclip helps people manage AI agents and their work.
> - Agent chat shows which agent receives each message.
> - The chat header used initials instead of the agent avatar.
> - The name and controls did not use the same center alignment.
> - This change uses the shared avatar and centers the header content.
> - Users can identify the agent and open its settings from the same
row.
## Linked Issues or Issue Description
**What happened?**
The agent chat header showed initials instead of the agent avatar. The
settings control did not align with the name and avatar.
**Expected behavior**
Show the agent avatar. Put the avatar, name, and settings control on the
same horizontal center line.
**Steps to reproduce**
1. Enable Agent Chat in Experimental settings.
2. Open a conversation with an agent that has an appearance set.
3. Check the avatar and settings control in the top bar.
Related navigation work: #14706.
## What Changed
- Use `AgentAvatar` in the conversation breadcrumb.
- Update the breadcrumb key when the agent appearance changes.
- Center breadcrumb labels that have an adjacent action. Keep task
identifier baseline alignment unchanged.
- Add regression checks for avatar props, appearance changes, and center
alignment.
## Verification
- PASS: affected Vitest suites, 129 tests.
- PASS: `pnpm check:token-gates`.
- PASS: `pnpm --filter @paperclipai/ui typecheck` on an isolated retry.
- PASS: `pnpm --filter @paperclipai/ui build`.
- PASS: browser checks at 1000 and 390 CSS pixels. Avatar, name, and
settings control all have center Y = 29.5 CSS pixels.
- Screenshots use the real header and avatar renderer with isolated
context fixtures. No screenshot or fixture is part of this diff.
- Full typecheck and build cannot complete because Cargo is not
installed.
- Full tests stopped with SIGKILL. The first UI typecheck also stopped
with exit 137. These runs do not prove full-suite success.
## Risks
- Low risk. The change only affects UI rendering. It changes no API or
database contract.
- Breadcrumbs with adjacent actions now use center alignment. Ordinary
task identifiers keep baseline alignment.
## Model Used
OpenAI Codex agent, with code editing, shell tools, and browser checks.
The runtime did not expose the exact model ID or context-window size.
## 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)
- [ ] 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
No behavior or command documentation needs an update for this rendering
fix. The execution environment requires the assigned branch name to stay
unchanged. Pending review gates are not marked complete.
Co-authored-by: Paperclip <noreply@paperclip.ing>
## Thinking Path
> - Paperclip is the open source app people use to manage AI agents for
work.
> - Agents reach external services through the Apps catalog. Each
catalog entry is a reviewed `AppDefinition` that wires a provider's
hosted MCP server into Paperclip's shared vault, grants, policies,
gateway, and audit trail.
> - Neon is a widely used serverless Postgres provider with an official
hosted MCP server, but it is not in the catalog. Teams that run their
databases on Neon must use the generic "connect your own MCP server"
path, which has no branding, no guidance, and no project or read-only
controls.
> - The connector playbook requires a catalog entry for a provider like
this: the hosted server supports dynamic client registration and bearer
API keys, and the common definition fields can express every option
Paperclip can serialize.
> - This pull request adds the Neon definition, its official artwork,
the research and permission-review ledger rows, documentation, and
deterministic tests, without any provider-specific runtime code.
> - The benefit is a one-click, governed Neon connection with optional
project pinning and read-only mode, and a documented path to live
qualification.
## Linked Issues or Issue Description
**Problem or motivation**
Neon is a common Postgres host for the applications agents work on, but
Paperclip's Apps catalog has no Neon entry. Operators who want agents to
inspect schemas, run SQL, or manage branches must paste the MCP URL into
the generic remote-MCP flow, which gives no branding, no provider
guidance, no project boundary, and no read-only switch.
**Proposed solution**
Add a catalog-only Neon connection built from the connector playbook:
browser sign-in through Neon's dynamic client registration with the
reviewed `read` and `write` scopes, or a customer API key sent as an
Authorization bearer header. Both methods expose Neon's documented
`projectId` pin and `readonly` switch as optional Advanced fields. Every
discovered tool stays governed by the normal per-action policies.
**Alternatives considered**
A plugin was not needed because no custom UI, tables, workers, or
webhooks are involved. A separate read-only method was not added because
the playbook treats read-only switches as advanced fields rather than
methods. Neon's repeatable `category` query filter was left out because
tenant fields serialize lists as one comma-joined value, so it cannot be
sent correctly without new runtime code; per-action policies cover
catalog narrowing instead.
**Roadmap alignment**
This extends the existing self-serve remote-MCP connection catalog and
does not overlap planned core work.
## What Changed
- Added the `neon` provider to `scripts/ingest-app-definitions.mjs`
(category, API-key placement, methods, tenant fields, guidance,
warnings) and regenerated
`packages/shared/src/app-definitions/neon.json` plus the generated
registry.
- Added the Neon row to the self-serve MCP research ledger with
`dcr_or_api_key` auth and risk tier S4.
- Added permission reviews for `neon/mcp-oauth` (explicit scopes `read`,
`write`, taken from Neon's live authorization-server metadata) and
`neon/mcp-api-key` (provider key), with evidence links.
- Added Neon's official tile icon (`ui/public/brands/apps/neon.png`,
copied byte-for-byte from the icon linked by neon.com) and the brand
manifest entry.
- Added prosumer gallery copy for the Neon card.
- Added `doc/connections/NEON.md` (service involvement, endpoints,
administrator setup, capabilities and policy, manifest, brand
provenance, validation hook) and linked it from the connections README
and the permission audit.
- Tests: Neon definition shape, store visibility and artwork, URL
recognition, reviewed scopes with scope-widening rejection, URL
projection of the project pin and read-only flag, invalid project ID
rejection, the connect form's API-key gating, and the pinned catalog
counts.
## Verification
- `pnpm exec vitest run packages/shared/src/app-definitions.test.ts
packages/shared/src/app-definitions-url.test.ts` — 34 passed.
- `pnpm exec vitest run
server/src/__tests__/tool-access-service.test.ts` — 367 passed.
- `pnpm exec vitest run ui/src/pages/apps/AppsConnect.test.tsx
ui/src/pages/apps/Browse.test.tsx ui/src/lib/app-brand-assets.test.ts
ui/src/pages/apps/AppLogo.brand-assets.test.tsx` — all passed.
- `node scripts/check-app-brand-assets.mjs` and `node --test
scripts/app-brand-validation.test.mjs` — passed.
- `pnpm --filter @paperclipai/shared typecheck`, `pnpm --filter
@paperclipai/server typecheck`, `pnpm --filter @paperclipai/ui
typecheck`, `pnpm check:token-gates` — clean.
- Manual: in a local instance, open Apps → Browse, confirm the Neon card
and icon, open `/apps/connect?source=neon`, confirm both methods, the
Advanced project pin and read-only toggle, and that Connect enables
after an API key is entered. The operator completed a live connection
against a Neon account on this build.
- Live metadata probed on 2026-10-02: both `.well-known` documents at
`mcp.neon.tech` return the recorded endpoints and scopes; an
unauthenticated `initialize` returns 401 with `resource_metadata`.
## Risks
- Low risk to existing providers: the change is additive catalog data
plus tests. The generated registry only gains one import.
- Neon's hosted server grants broad project and database management. The
definition carries two warnings, recommends a development project, and
keeps every write under the normal action policies; the read-only switch
is enforced by Neon's server, not locally.
- The permission-review ledger records live proof for both methods as
not run; the full lifecycle checklist in `doc/connections/NEON.md` still
needs a documented pass before the entry is considered fully qualified.
> 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
- Claude Fable 5.1 (`claude-fable-5-1`) in Claude Code, with extended
thinking and tool use (shell, file editing, browser verification).
## 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>
Use stock discovery metadata and an actual Markdown document-link example. Clarify the future explicit UI-link fixture and retain the original task request and all measured failures.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Include the issue-document reference in the normative inventory and calibrate coverage of every declared source without changing each inventory’s existing heading parsing scope.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Regenerate both capability inventories for the early issue-document recipe and reject stale derived manifests before stock-harness provider admission. Preserve the eight measured repair results, automatic recovery cost, and OpenCode delivery diagnosis.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Preserve the tiny hire manual and original assigned-skill case. Add a focused public document/revision/link oracle and explicit presence/absence provenance for a matched skill-only comparison.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Run credential-free gates before live admission and verify retained evidence in direct browser execution. Include evaluated instruction sources in suite revisions and calibrate stale/missing evidence rejection.
Co-Authored-By: Paperclip <noreply@paperclip.ing>
## Thinking Path
> - Paperclip is the open source app people use to manage AI agents for
work.
> - The heartbeat service records each agent turn and releases its
working files.
> - A turn can save its work and finish before instruction-copy cleanup
runs.
> - A cleanup exception can replace that completed result with an
adapter failure.
> - This also loses result accounting and can prevent environment lease
release.
> - This pull request records a cleanup warning and keeps the original
run outcome.
> - The existing recovery sweep retries cleanup from the durable
working-copy record.
## Linked Issues or Issue Description
Related cleanup and lock work: #14866 and #14869. Related, distinct
work: #14695 retains warm-process files; #12021 handles provider-process
SIGTERM after a terminal result.
**What happened?**
An agent saved its plan, posted a comment, and requested approval. The
provider completed its turn. Instruction-copy cleanup then timed out on
a directory lock. Its exception escaped a `finally` block and replaced
the provider result, so the completed turn showed `Run failed`.
**Expected behavior**
Keep the provider outcome, usage, cost, saved work, and pending
approval. Record a cleanup warning and let the existing recovery sweep
retry. A real provider failure must keep its original error. A failed
file save must keep its failed-save receipt.
**Steps to reproduce**
1. Complete a legacy adapter turn that saves work and requests approval.
2. Make instruction-copy release throw a directory-lock timeout.
3. Read the run result. Before this change, the cleanup error replaces
the provider outcome.
The new heartbeat tests reproduce the failure without a live provider or
external service.
**Paperclip version or commit**
The regression reproduces on `c83df091b1a5207375eaf23466bb5c62e4e1518e`.
This branch is rebased onto `cf8ad63c80`.
**Deployment mode**
Server-managed agent execution with persistent instruction working
copies.
## What Changed
- Catch instruction-copy release failures in both heartbeat teardown
paths. Stop repeating a failed cleanup attempt within the same run.
- Write a sanitized `instruction_cleanup` warning. A warning-write
failure also preserves the run result.
- Test successful, failed, and throwing providers; both teardown paths;
warning-write failure; accounting; approval state; and execution-control
release.
- Extend the held-lock test to prove a fresh recovery worker removes the
deferred copy and preserves its failed-save receipt.
- Document deferred cleanup and the run-log event.
## Verification
- Red proof: all five new heartbeat regression cases fail with the
original release calls.
- At head `20bea4f431c916d2f5db1970213aab85f5daa34c`, all 417 tests
passed across heartbeat process recovery, agent directory working
copies, and directory merge locks.
- Full local `pnpm -r typecheck`, `pnpm build`, and `git diff --check`
passed.
- [GitHub
CI](https://github.com/paperclipai/paperclip/actions/runs/37031119795)
passed at this head. All 53 reported checks passed; the two Storybook
checks were correctly skipped. This includes general and serialized
tests, browser shards, runner verification, build, typecheck, and the
canary dry run.
- Greptile reviewed this head with 5/5, no code comments, and no
unresolved review threads. The branch has no merge conflicts.
- The local `pnpm test:run` attempt was stopped after it reported eight
failures in unchanged suites. Four Slack/AgentMail cases selected an
unrelated ancestor skills directory and failed with `ENOENT`; the two
Slack cases passed with a temporary local skill-root link, which was
then removed. Three company-skill cases reproduced macOS `EACCES` errors
when renaming read-only cache directories. One gateway case passed when
rerun alone. No full local-suite pass is claimed; the complete CI test
jobs passed.
## Risks
- Cleanup errors now leave recovery work pending. The durable
working-copy record remains available for the existing retry sweep.
- This change preserves provider failures and failed-save receipts. It
does not claim that unsaved file edits were saved.
- Native instruction reservation errors retain their existing behavior
because they guard process ownership.
- No schema, lockfile, workflow, API, or UI changes.
## Model Used
OpenAI Codex, based on GPT-6, with reasoning, repository tools, and code
execution. The exact backend model ID and context-window size are not
exposed by this session.
## 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 open source app people use to manage AI agents for
work.
> - Each agent runs on an adapter, and the operator picks the agent's
model from the list the adapter advertises.
> - The `codex_local` adapter advertises a static, curated list. The
server returns that list as written, but the list itself is not in a
useful order: the default `gpt-5.6-sol` sits above the whole GPT-6
family (#14878).
> - This pull request orders the list the way the ChatGPT app orders
Codex models: newest model version first, then by decreasing capability
inside each version, with older models at the end.
> - It is stacked on #14917, which makes the model dropdown show a
hand-ordered list as the adapter advertises it. Without that change the
picker shows every list alphabetically.
> - The benefit is that a user who knows Codex finds the right model at
once, and older models sit at the end of the list.
## Linked Issues or Issue Description
Fixes#14878.
Related: #14917 (the Claude counterpart, #14877) carries the dropdown
change this PR relies on. This PR contains that commit until #14917
lands; after that it rebases to the Codex commit alone.
## What Changed
- `packages/adapters/codex-local/src/index.ts`: reorder the advertised
`models` list. GPT-6 (`astra`, `sol`, `luna`) first, then GPT-5.6
(`sol`, `terra`, `luna`), then `gpt-5.5`, `gpt-5.4`, `gpt-5.4-mini`,
`gpt-5`, `gpt-5-mini`, `gpt-5-nano`, then the o-series and
`codex-mini-latest` in their existing relative order.
`DEFAULT_CODEX_LOCAL_MODEL` is unchanged.
- `packages/adapters/codex-local/src/index.test.ts`: the metadata test
now asserts the full order of the GPT entries.
The ChatGPT app also lists GPT-6.1 Sol first. That model is not in
Paperclip's list today. Adding a model needs reasoning-effort and
fast-mode entries as well, so it is out of scope for an ordering fix.
## Verification
Run from the repository root:
```sh
pnpm exec vitest run packages/adapters/codex-local server/src/__tests__/adapter-models.test.ts
pnpm --filter @paperclipai/adapter-codex-local typecheck
```
Red on `master`, green here: the updated metadata test fails against the
unmodified list because `gpt-5.6-sol` comes first.
Manual check: open an agent that uses `codex_local` and open the model
dropdown. The list reads gpt-6-astra, gpt-6-sol, gpt-6-luna,
gpt-5.6-sol, gpt-5.6-terra, gpt-5.6-luna, gpt-5.5, gpt-5.4,
gpt-5.4-mini, gpt-5, gpt-5-mini, gpt-5-nano, o3, o4-mini, o3-mini, Codex
Mini.
## Risks
- Low risk. The list content and the default model do not change; only
the order does. The server returns this list as-is for the built-in
Codex adapter and the server test compares against the exported list, so
no server change is needed.
- The first entry is no longer the default model. Nothing reads the
first entry as a default: `DEFAULT_CODEX_LOCAL_MODEL` is resolved
separately.
## Model Used
Anthropic Claude Fable 5.1 (`claude-fable-5-1`) through Claude Code,
extended thinking on, with tool use for reading the repository, running
vitest and tsc, and editing files. The account holder reviewed the
change and owns the commit.
## 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
---------
Co-authored-by: abderbj <115119179+abderbj@users.noreply.github.com>
## Thinking Path
> - Paperclip is the open source app people use to manage AI agents for
work.
> - Each agent runs on an adapter, and the operator picks the agent's
model from a list the adapter advertises.
> - The `claude_local` adapter advertises a static list and merges in
the models the Anthropic API returns. Neither list has a deliberate
order.
> - The model dropdown then sorts every list by id. For Claude this
shows "Fable 5", "Fable 5.1", "Haiku 4.5", "Mythos 5", "Opus 4.6" ...
which is not the order of capability, release, or version (#14877).
> - This pull request gives the adapter one defined order, the one the
Claude app uses: the newest release of each family first, by decreasing
capability, then older releases grouped by family. The server applies it
to discovered models, and the dropdown keeps the adapter's order instead
of re-sorting.
> - The benefit is that a user who knows the Claude app finds the right
model at once, and older models sit at the end of the list.
## Linked Issues or Issue Description
Fixes#14877.
Related: #14147 touches the same adapter's model list (the
`ANTHROPIC_MODEL` default label) and does not change ordering. #14878 is
the Codex counterpart and depends on the dropdown change in this PR.
## What Changed
- `packages/adapters/claude-local/src/server/model-order.ts` (new):
`sortClaudeModels()` and `parseClaudeModelId()`. The parser reads the
current scheme (`claude-opus-4-8`), the legacy scheme
(`claude-3-7-sonnet-20250219`), dated snapshots, `-latest` aliases, the
`[1m]` suffix, and Bedrock ids (`us.anthropic.…-v1`, `…-v2:0`). The sort
puts the newest release of each family first (Fable, Mythos, Opus,
Sonnet, Haiku), then older releases grouped by family with versions
descending. An alias sorts before its dated snapshots, and dated
snapshots of one release sort newest first. Ids that are not Claude
models keep their incoming order at the end.
- `packages/adapters/claude-local/src/server/models.ts`: apply the order
to the static fallback, to the merged API list, and to the Bedrock list.
Reorder `BEDROCK_MODELS` to match.
- `packages/adapters/claude-local/src/index.ts`: reorder the advertised
`models` list to the same order.
- `ui/src/components/AgentConfigForm.tsx`: `ModelDropdown` gets a
`preserveOrder` prop. With it the dropdown shows the list as the adapter
ordered it; without it the list is sorted by id as before.
`ui/src/lib/model-utils.ts` adds `adapterCuratesModelOrder()`, true for
the built-in adapters whose list arrives in a deliberate order
(`claude_local`, `codex_local`, `paperclip_runner`, `gemini_local`,
`grok_local`, `kimi_local`, `openclaw_gateway`, and `opencode_local` /
`pi_local`, which the server sorts when discovered and which lead with
the default model when it falls back to the declared list). Cursor is
not in the set because its list comes from `agent models` discovery, and
adapters not named there, including externally installed ones, keep the
alphabetical fallback. The three dropdown call sites (`AgentConfigForm`,
`ConfigureBuiltInAgentModal`, `NewAgentSetup`) pass it; `NewAgentSetup`
decides by the resolved brand type, because a `paperclip_runner` agent
fetches the Claude or Codex list for its brand. Grouped lists
(`opencode_local`, `pi_local`) are unchanged.
- Tests: `model-order.test.ts` (adapter, including the snapshot-date
tie-breaker), `ModelDropdown.test.tsx` (ui: preserved order with the
prop, alphabetical without it, provider groups unchanged),
`model-utils.test.ts` (which adapters opt in), and two updated
expectations plus one new order assertion in
`server/src/__tests__/adapter-models.test.ts`.
Mythos is not in the Claude app's list. This PR ranks it directly after
Fable, in the top capability tier. The rank table in `model-order.ts` is
one line to change if you prefer a different slot.
## Verification
Run from the repository root:
```sh
pnpm exec vitest run packages/adapters/claude-local server/src/__tests__/adapter-models.test.ts ui/src/lib/model-utils.test.ts ui/src/components/ModelDropdown.test.tsx ui/src/components/AgentConfigForm.render.test.tsx ui/src/components/ConfigureBuiltInAgentModal.test.tsx ui/src/pages/NewAgent.test.tsx
pnpm --filter @paperclipai/adapter-claude-local typecheck
pnpm --filter @paperclipai/ui typecheck
pnpm --filter @paperclipai/server typecheck
pnpm check:module-boundaries && pnpm check:token-gates && pnpm check:tokens
```
Red on `master`, green here:
- `ModelDropdown.test.tsx` fails against the unmodified dropdown because
the ids come back sorted alphabetically even with `preserveOrder`.
- `adapter-models.test.ts` fails against the unmodified adapter because
the first model is `claude-opus-4-8`, not `claude-fable-5-1`.
Manual check: open an agent that uses `claude_local`, open the model
dropdown. With no `ANTHROPIC_API_KEY` the list reads Fable 5.1, Mythos
5, Opus 5.5, Sonnet 5, Haiku 4.5, Fable 5, Opus 5, Opus 4.8, Opus 4.7,
Opus 4.6, Sonnet 4.6, Sonnet 4.5. With a key, the discovered models slot
into the same order.
## Risks
- The other built-in adapters in the set (Gemini with `Auto` first,
Grok, Kimi, OpenClaw, the runner's Codex list, and the OpenCode and Pi
lists in the built-in-agent modal) are now shown as their adapter
delivers them instead of alphabetized. Cursor's discovered list and
every adapter outside the set, including externally installed ones, keep
the alphabetical order they had, so no option moves between refreshes.
Grouped lists are unchanged.
- The first entry of the Claude list changes from Opus 4.8 to Fable 5.1.
Nothing reads the first entry as a default: `DEFAULT_CLAUDE_LOCAL_MODEL`
is `claude-opus-5` and is resolved separately.
- No API, schema, or migration change.
## Model Used
Anthropic Claude Fable 5.1 (`claude-fable-5-1`) through Claude Code,
extended thinking on, with tool use for reading the repository, running
vitest and tsc, and editing files. The account holder reviewed the
change and owns the commit.
## 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
---------
Co-authored-by: abderbj <115119179+abderbj@users.noreply.github.com>
## Thinking Path
> - Paperclip is the open source app people use to manage AI agents for
work.
> - Agents use tools through company-scoped connections and provider
sessions.
> - Resolving a tool connection currently forces a fresh session even
when the provider can load new tools into the existing conversation.
> - A fresh provider conversation can receive too little history to
continue the task.
> - This pull request adds explicit tool-refresh capabilities and uses
them in both runner paths.
> - Fresh attempts receive bounded task history with source IDs and
retrieval instructions.
> - The benefit is that agents can continue the same task after a
connection changes.
## Linked Issues or Issue Description
**What happened?**
A resolved tool connection forced a fresh provider conversation. The new
conversation could lose the original goal and prior answers. Claude also
rejected resume when only the MCP server set changed.
**Expected behavior**
Resume the provider conversation when its harness can refresh tools.
When a fresh session is required, supply enough bounded history to
continue the task. Preserve company, agent, task, workspace, model,
instruction, and skill checks.
**Steps to reproduce**
1. Start a conversation and agree on a task and its constraints.
2. Request and connect a tool needed for the task.
3. Continue the conversation after the connection resolves.
4. Check that the agent remembers the task and can use the new tool.
**Paperclip version or commit**
The bug was reproduced on master at `c46e41e81`. This branch is rebased
on current master.
**Deployment mode**
Self-hosted server. Both legacy adapters and the native runner are
affected.
Related public work: Refs #13282 for task-backed conversations. Refs
#13057 for the broader session-compaction proposal. Refs #14659 for
another report about local CLI session continuity. This change fixes
tool-connection continuation. Provider authentication repairs keep their
existing recovery behavior.
## What Changed
- Expose tool-refresh support in native harness descriptors and legacy
adapter metadata.
- Request tool refresh after connection resolution. Keep provider
authentication repair as a fresh-session wake.
- Reload current tools and credentials while retaining supported Claude,
Codex, Grok, and other provider conversations.
- Allow MCP-only changes during qualified native recovery. Keep all
other compatibility checks.
- Refresh managed-provider and ACPX tool bindings when attaching a new
run.
- Add a fresh-session handoff for both runner paths. Bound database
reads, excerpts, and the final packet to 24,000 bytes.
- Include the original request, recent messages, decisions, plans, prior
answers, and source IDs. Mark omitted content. Apply reset boundaries,
wake cutoffs, quarantine, and secret redaction.
- Add regression tests and document the capabilities and handoff
behavior.
## Verification
- `pnpm -r typecheck` and `pnpm build` passed. Rust formatting passed.
- Final review fixes passed 314 server tests, 333 adapter utility tests,
139 native-session runtime tests, and 12 managed-provider Rust tests.
They verify historical quarantine, raised budgets across attachment, no
history reads on successful resume, and handoff delivery on fresh retry.
- Broader branch verification also passed 1,401 adapter utility tests,
1,047 runner TypeScript tests, 43 Grok adapter tests, and 311 Rust core
tests.
- Live Claude CLI and Grok ACP probes preserved the provider session ID,
recalled a prior task constraint, and called a newly added read-only MCP
tool.
- GitHub CI passed on `b21486d18084a7aa4cafbe8e012f7cad6585d9cc`: 55
successful checks and 4 skipped checks. This includes all test shards,
all eight browser shards, runner checks, and the Grok clean public npm
install canary. [CI
run](https://github.com/paperclipai/paperclip/actions/runs/37057514976).
- A full local test attempt encountered a separate Git snapshot timeout.
All affected local suites passed after the final edits, and the full CI
test gates passed.
- Review the capability matrix in `packages/paperclip-runner/README.md`.
Repeat the four reproduction steps with a supported provider and with an
unsupported harness.
## Risks
- Provider tool refresh can fail. Existing recovery falls back to a
fresh conversation where policy permits it.
- A new transport can replace an old process while preserving the
provider conversation. Tests cover current credentials and unchanged
identity.
- Long history can omit older context. Explicit markers and source IDs
let the agent retrieve needed context within task scope.
- Unknown and unqualified harnesses use the fresh-session path. No
database migration is required.
## Model Used
OpenAI Codex, GPT-6, with reasoning, tool use, code execution, and live
provider testing. The exact model ID and context-window size are not
exposed in this session. Claude and Grok also ran as test subjects.
## 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 open source app people use to manage AI agents for
work.
> - Slack chat connections let people start and continue agent work from
Slack.
> - The setup prompt guides an agent through the Paperclip and Slack
browser interfaces.
> - A live setup completed, but several instructions did not match the
current interfaces.
> - Those gaps can send users to the wrong connection flow or leave them
waiting for controls that do not appear.
> - This pull request updates the prompt with the steps observed during
the live setup.
> - The benefit is a clearer path from a fresh instance to a verified
Slack conversation.
## Linked Issues or Issue Description
Refs: #13920 and #14862. The first added the Slack conversation flow.
The second updated the shared setup prompt control. A search found no
duplicate open PR or matching public issue.
**Issue type**
Outdated instructions and missing setup guidance.
**Where is the issue?**
`ui/src/pages/apps/chat/SlackSetupPrompt.tsx`
**What's wrong?**
The prompt omits the Chat connectors setting on fresh instances. It uses
an old navigation label. It assumes an avatar crop dialog and a Save
button always appear. It also assumes the suggested bot username matches
Slack and that the Slack reply contains a task link.
**Suggested fix**
Use the current labels. Explain the feature prerequisite, avatar save
behavior, real mention selection, clipboard recovery, and the path to
task and run evidence.
## What Changed
- Add the Chat connectors prerequisite and current Connectors, resume,
and identity-link labels.
- Explain Slack's combined Create and Install action and how to continue
from its success page.
- Handle avatar uploads that save immediately. Require a reload to
confirm the saved icon.
- Select the real bot from Slack's mention suggestions, including names
with punctuation.
- Recover from an empty or stale clipboard without exposing credentials.
- Find the linked task through Conversations and inspect the agent's
Runs page.
## Verification
- `pnpm exec vitest run
ui/src/pages/apps/chat/SlackSetupPrompt.test.tsx`: 10 tests passed after
rebasing onto current master.
- `git diff --check origin/master...HEAD`: passed.
- `pnpm build`: passed.
- `pnpm -r typecheck`: passed.
- `pnpm check:token-gates`: passed.
- Full Vitest suite: passed in CI for this commit, including all server,
chat, workspace, and serialized test shards. The duplicate local `pnpm
test:run` was stopped after CI finished; it did not complete locally.
- Current-commit CI: all checks passed, including build, typecheck, E2E,
Runner verification, and canary dry run. Greptile rated the change 5/5
with no findings or unresolved review threads.
- Live browser test before the wording update: created and installed a
new Slack app, verified the callback, uploaded and reopened the avatar,
linked the configuring user's identity, and enabled the selected test
channel. The first mention received a reply. A follow-up without another
mention recalled the first message. Both agent runs succeeded.
- The wording update does not repeat Slack app installation. Existing
tests verify the complete copied prompt, clipboard fallback, and
instance URL handling.
- This is a prompt-text refactor. No runtime behavior changes, so the
existing tests cover the copied result without a new test that repeats
the wording.
## Risks
- Low risk. This change updates prompt text in one file.
- Provider interfaces can change. The prompt tells the agent to inspect
the current page and handle optional controls.
- The prompt handles the observed mention mismatch. This change does not
alter the generated suggested username.
## Model Used
- OpenAI Codex, based on GPT-6, with reasoning, repository tools, code
execution, and browser automation. The session does not expose an exact
runtime model ID or context window size. The live test also used an
earlier model whose exact ID was not exposed.
## 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>
