Commit Graph
2 Commits
Author SHA1 Message Date
DottaandClaude Opus 5.5 b56f240f45 feat(analytics): show session replay text by default, mask only customer content (#22057)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Hosted Paperclip Cloud records browser sessions with PostHog session
replay
> - Replay masked all text and showed only an allowlist of chrome, so
most replays showed only stars
> - The team cannot see how people use the board, and the allowlist
needed a review for each new label
> - The owner decided to show text by default and mask only marked
customer content, the way PostHog normally works
> - This pull request flips the default, removes the allowlist, and
marks every surface that can show connected-account content
> - The benefit is readable replays, while Google user data and Slack
data stay masked

## Linked Issues or Issue Description

**Subsystem affected**

Hosted Cloud browser analytics: PostHog session replay in the board UI
(`ui/src/lib/analytics/`).

**Problem or motivation**

Replay masked every text node. Only sidebar items and opt-in components
showed text. Replays were almost unreadable. Each new label needed an
`analyticsChrome` opt-in and an allowlist review.

**Proposed solution**

Record text as it is. Mask text only inside `ph-mask` (customer
content), `ph-no-capture` (secrets, blocked) and editable content. Add
`ANALYTICS_MASK_CLASS` to every surface that can show connected-account
content (Gmail, Calendar, Drive and Docs, Slack) or other free-form text
from agents or customers.

**Alternatives considered**

Keep "mask all" and grow the allowlist. This needs a review for each
label and still hides most of the board.

## What Changed

- `privacy.ts`: `ANALYTICS_MASK_TEXT_SELECTOR` is now `.ph-mask,
.ph-no-capture, [contenteditable]:not([contenteditable="false"])`. rrweb
masks a text node only when its parent or an ancestor matches the
selector. Removed `maskReplayText`, `isReplayTextRevealed`,
`ANALYTICS_UNMASK_ATTRIBUTE` and `ANALYTICS_UNMASK_PROPS`.
- `product-analytics.ts`: removed `maskTextFn`, so rrweb uses its
default star masking. `maskAllInputs`, `blockClass`, `maskAttributeFn`,
`mask_all_text` and `mask_all_element_attributes` do not change.
- Removed the `analyticsChrome` prop from `Button`, `Label`,
`TabsTrigger`, `EmptyState` and `SidebarNavItem`, and from all sidebar
call sites.
- Added `ANALYTICS_MASK_CLASS` to the content element (not the chrome)
of these surfaces:
- Link preview cards, task quicklook description, external object pills
and rows, related-work titles.
  - Email message cards and email send intents.
- Connected account names and email addresses (Browse, Connections, app
detail headers, app sidebar, connection choice list, connection access
and repair cards), GitHub login and repositories, Slack workspace names,
chat channel and linked user labels.
- Inbox and decision rows (subject and detail), obligation evidence,
Needs You notes and receipts, decision options and result summaries, and
the interaction card roots (questions, options, answers, tool action
requests, suggested tasks).
- Approval payloads, action card input and reason, recovery card
headline, evidence and options, run ledger retry detail, interrupt
handoff previews.
- Run transcripts, run failure details and run events on the agent page.
- Search result snippets, toast bodies, voice and call transcripts, case
summary, files and fields, attachment previews, artifact previews (text,
tables, patches and documents).
- Replaced `unmask-allowlist.test.ts` with `replay-masking.test.tsx`. It
checks that unmarked chrome records as it is, that each marker masks,
that a rendered voice transcript and run transcript mask, that each
surface above has the class, and that autocapture records no element
text. Updated `privacy.test.ts`, `shared-chrome.test.tsx` and
`product-analytics.test.ts`.
- Updated `doc/cloud-browser-analytics.md`: new default, marker classes,
masked surfaces, and residual risk.

## Verification

- `pnpm exec vitest run src/lib/analytics` in `ui/`: 83 tests pass.
- Vitest for the 48 test files next to the changed components: 994 tests
pass.
- `tsc --noEmit -p ui/tsconfig.json`: no new errors (only the existing
generated Storybook catalog errors).
- `pnpm verify` exits 2 because a doc is a global CI input. Required CI
supplies the broad coverage.
- Manual: I ran an isolated local instance with a Cloud-like browser
analytics config. `PAPERCLIP_POSTHOG_HOST` pointed to a local capture
server, not PostHog. I opened a task page in headless Chromium, hovered
a GitHub pull request link, and decoded the captured `/s/` replay
payloads:
- Clear text: sidebar (`New task`, `Chat`, `Tasks`, `Inbox`, `More`),
breadcrumb, task title `Quarterly launch checklist`, `REP-1`, the
Connectors promo heading and button.
- Masked (stars only): the markdown task body, the chat comment, the
link preview card (pull request title, author, URL, status), and the
company menu.
- Marker strings in the body, comment and preview never appeared in
clear text in any payload.
- The page `<title>` (task title and company name) is in the replay
head. This is accepted residual risk.

## Risks

- Text that is not marked is now recorded. Task titles, agent names,
project names, company names, routine names, artifact and document
titles, and activity rows are visible in replays. The owner accepted
this. An agent can copy email or Slack text into a task title.
- A new surface that shows connected-account content without
`ANALYTICS_MASK_CLASS` is recorded until somebody marks it.
`replay-masking.test.tsx` and the doc list the marked surfaces.
- The masking options in `posthog.init` take precedence over the PostHog
project masking settings (posthog-js 1.450), so a dashboard change
cannot unmask content. Keep the project settings empty to avoid a
console warning.
- Some masked areas also mask nearby fixed labels (for example the
interaction card buttons). This is safe.
- Analytics is off on self-hosted and local installs, so they see no
change.

> 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 Opus 5.5 (claude-opus-5-5), Anthropic, with tool use and code
execution in Claude Code.

## 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

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-17 15:18:50 -06:00
Devin FoleyandPaperclip 98e3cec78e feat(costs): report monthly subscriptions beside API spend (#15634)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Its Costs page reports API charges and subscription token usage.
> - Subscription runs do not show the recurring fee paid for the
account.
> - Several agents can share one paid seat, so a fee per agent would
overstate cost.
> - This pull request identifies connected subscription accounts and
reports each monthly fee once.
> - Users can compare API spend and subscription costs, see uncertainty,
and enter their actual price.

## Linked Issues or Issue Description

Refs #498 and #3406 for related subscription reporting needs. This
change does not add subscription budget enforcement or automatic
optimization.

Related prior approaches: #10544 and #11104 proposed API-equivalent
token costs for subscription work. This PR reports recurring account
fees separately from usage charges and budgets. A search found no open
PR for monthly account fee reporting.

**Subsystem affected**
Costs, managed AI connections, shared contracts, database schema, and
board UI.

**Problem or motivation**
An organization can pay for subscriptions even when included runs record
no additional dollar charge. The Costs page cannot show that monthly
commitment or distinguish a shared paid seat from several separate
seats.

**Proposed solution**
Discover account identity from the managed connection selected for a
run. Use exact known plans for labeled monthly list-price estimates.
Keep ambiguous plans and unconfirmed identities unpriced until the user
supplies a price. Sum current monthly fees separately from date-filtered
API charges and token usage.

**Alternatives considered**
A fee per agent double-counts shared accounts. Token-based API price
estimates do not measure the recurring subscription fee. Provider
invoices would be more exact but are not available through these account
credentials.

**Roadmap alignment**
Extends the shipped Better Budgeting and Personal & Shared AI Accounts
capabilities. The feature and UI were reviewed with the maintainer
before this PR.

## What Changed

- Add company-scoped subscription accounts, connection bindings, and
immutable price revisions in replay-safe migration 0322.
- Identify OpenAI workspace/seat and Claude organization/account pairs
only after provider authentication. Keep provisional records
grant-scoped; unchecked ID-token claims cannot combine accounts or grant
billing editors. Keep credentials and raw provider identities out of
reports.
- Attach the selected subscription to new run receipts. Resolve
duplicate aliases without rewriting old monetary receipts.
- Add cached background discovery with four provider checks and twenty
discovery requests at a time. Accepted requests wait for capacity;
excess requests receive HTTP 429 and a UI retry action. Preserve saved
prices and observations when a provider check fails. Dashboard polls
load current prices only; audit revisions stay in the database.
- Support owner-controlled monthly or annual prices, fractional-cent
precision, separate currency totals, tracking status, and duplicate
linking. Preserve both personal owners and shared-account managers when
discovery finds the same paid seat.
- Show API spend beside monthly subscription commitment, with separate
token totals and clear estimate labels. Account details and fee totals
follow billing-owner/shared-audience visibility; company-wide token
totals remain available to cost readers.
- Use consistent amount and estimate badge formatting for agent, user,
project, and model rows.
- Add real PostgreSQL tests, route authorization tests, UI tests,
interactive Storybook fixtures, and operator documentation.

## Verification

- Current head `d7b5c78dadf5033a3c447d83434676ac94687e77` passed all 53
checks, with two checks skipped. This includes the full test matrix,
browser tests, build, typecheck, migration policy, canary packaging, and
security checks.
- Greptile Apex reviewed that exact head at 5/5 with no unresolved
findings.
- Local `pnpm -r typecheck`, `pnpm build`, `pnpm check:token-gates`,
migration safety checks, and `git diff --check` passed.
- All 238 focused tests passed across accounting, subscription
persistence, authorization, UI, migration replay, and snapshot drift.
- The expanded local chat suite passed 1,149 tests. Two cases failed
because this workspace found an unrelated ancestor skill directory; both
passed unchanged in an isolated checkout. The full local Vitest command
was not repeated after the rebase because of this environment issue. CI
covered the complete test matrix on the final head.
- Migration `0322_reflective_kree.sql` follows master's
`0321_acoustic_human_fly.sql`. Its snapshot and journal were
regenerated. Master's migration and snapshot are unchanged. The
subscription SQL retains its original content and hash.
- Disposable PostgreSQL upgrade fixtures applied master's added
lifecycle and Slack migrations without changing saved subscription
prices or cost receipts. The renamed subscription migration was
recognized by hash, retained one history row, and a second migration run
was a no-op.
- Regression tests cover provider capacity, fractional-cent edits,
billing visibility after grant revocation, and
fixed-price/credit/unknown usage. The new UI billing-label tests fail
against the old wording and pass with the fix.
- Browser checks passed for desktop and mobile layouts, shared-account
totals, annual price edits, unknown prices, and stale provider data.
- A live OpenAI account probe returned HTTP 200 and a verified Pro
account identity. The generic Pro plan correctly remains unpriced.
- Live Claude discovery identified Max 20x and a $200/month estimate.
Two real Claude agent runs were attributed to that same account, with
two usage receipts, two agents, one monthly fee, and zero additional
usage charges. Two OpenAI connections also deduplicated to one account.
- Live Codex execution is not verified: both Paperclip's managed hello
test and a direct host `codex exec` fail with a workspace-routing
authentication error. The current credential passes the usage API probe.
This is separate from the successful account-discovery check.
- A local performance fixture with 50,000 usage events and 5,001 saved
price revisions produced a 1,589-byte report in 11.5 ms median / 12.2 ms
maximum over ten warmed runs. PostgreSQL used the subscription/revision
index to read one current price row. These are local measurements, not a
production latency guarantee.
- [Desktop, account details, and mobile
screenshots](https://pages.paperclip.ing/pr-15634-subscription-costs-d5d285e2/)
show the production Costs page with synthetic Storybook fixtures.
- To inspect the UI, run `pnpm storybook` and open **Costs /
Subscriptions**. Edits there change fixture data only.

## Risks

- Provider plan identifiers can change. Unknown plans remain unpriced.
Automatic prices are estimates before tax; discounts and negotiated
prices require a user-supplied amount.
- Subscription fees describe the current monthly commitment. They do not
follow the selected usage date range and do not consume agent budgets or
create finance events.
- Monthly fees cover only accounts visible to the viewer; company-wide
token totals retain their existing scope.
- Legacy or unmanaged usage can remain unattributed. Unconfirmed
identities require manual prices and may need explicit duplicate linking
after credential changes.
- Disconnecting a connection does not prove provider billing stopped.
Users must end or exclude tracking explicitly.
- Migration 0322 adds tables, a nullable cost-event column, and an
index. It is replay-safe, but normal migration locks still apply.

## Model Used

OpenAI Codex, based on GPT-6. This session did not expose a more
specific backend model ID or context-window size. Capabilities used:
code editing, shell execution, database tests, and 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>
2026-10-09 13:38:01 -07:00