- Make company_boundary test adversarial with cross-company stimulus
- Replace fragile not-contains:retry with targeted JS assertion
- Replace not-contains:create with not-contains:POST /api/companies
- Pin promptfoo to 0.103.3 for reproducible eval runs
- Fix npm -> pnpm in README prerequisites
- Add trailing newline to system prompt
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Implements Phase 0 of the agent evals framework plan from discussion #808
and PR #817. Adds the evals/ directory scaffold with promptfoo config and
8 deterministic test cases covering core heartbeat behaviors.
Test cases:
- core.assignment_pickup: picks in_progress before todo
- core.progress_update: posts status comment before exiting
- core.blocked_reporting: sets blocked status with explanation
- governance.approval_required: reviews approval before acting
- governance.company_boundary: refuses cross-company actions
- core.no_work_exit: exits cleanly with no assignments
- core.checkout_before_work: always checks out before modifying
- core.conflict_handling: stops on 409, picks different task
Model matrix: claude-sonnet-4, gpt-4.1, codex-5.4, gemini-2.5-pro via
OpenRouter. Run with `pnpm evals:smoke`.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Paperclip <noreply@paperclip.ing>
Scaffolds end-to-end testing with Playwright for the onboarding wizard.
Runs in skip_llm mode by default (UI-only, no LLM costs). Set
PAPERCLIP_E2E_SKIP_LLM=false for full heartbeat verification.
- tests/e2e/playwright.config.ts: Playwright config with webServer
- tests/e2e/onboarding.spec.ts: 4-step wizard flow test
- .github/workflows/e2e.yml: manual workflow_dispatch CI workflow
- package.json: test:e2e and test:e2e:headed scripts
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Migrate from single-bundle CLI publishing to publishing all @paperclipai/*
packages individually via Changesets. This fixes the "Cannot find package
@paperclipai/server" error when installing from npm.
Changes:
- Add @changesets/cli with fixed versioning (all packages share version)
- Make 7 packages publishable (shared, adapter-utils, db, 3 adapters, server)
- Add build scripts, publishConfig, and files fields to all packages
- Mark @paperclipai/server as external in CLI esbuild config
- Simplify CLI importServerEntry() to use string-literal dynamic import
- Add generate-npm-package-json support for external workspace packages
- Create scripts/release.sh for one-command releases
- Remove old bump-and-publish.sh and version-bump.sh
- All packages start at version 0.2.0
Usage: ./scripts/release.sh patch|minor|major [--dry-run]
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Combines version bump, build, publish, restore, commit, and tag into
a single ./scripts/bump-and-publish.sh command. Supports --dry-run.
Also restores cli/package.json to dev format after v0.1.1 publish.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Add esbuild config to bundle CLI with all workspace code for npm publishing
- Add build-npm.sh script that runs forbidden token check, type-check,
esbuild bundle, and generates publishable package.json
- Add generate-npm-package-json.mjs to resolve workspace:* refs to actual
npm dependencies for publishing
- Add version-bump.sh for patch/minor/major/explicit version bumping
- Add check-forbidden-tokens.mjs that scans codebase for forbidden tokens
(mirrors git hook logic, safe if token list is missing)
- Add esbuild as dev dependency
- Add build:npm, version:bump, check:tokens scripts to root package.json
- Update .gitignore for build artifacts
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Rename all workspace packages from @paperclip/* to @paperclipai/* and
the CLI binary from `paperclip` to `paperclipai` in preparation for
npm publishing. Bump CLI version to 0.1.0 and add package metadata
(description, keywords, license, repository, files). Update all
imports, documentation, user-facing messages, and tests accordingly.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add instructionsFilePath config to Claude and Codex adapters, allowing
agents to load external instruction files appended to the system prompt.
Claude uses --append-system-prompt-file; Codex prepends file contents
to the prompt. Add docs:dev script for local Mintlify preview.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace inline env vars in package.json dev scripts with a dedicated
node script that supports --tailscale-auth for private-network dev.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Document secret storage in DATABASE.md and DEVELOPING.md. Update
SPEC-implementation with company_secrets schema and indexes. Add
migrate-inline-env-secrets script for converting existing plain
env values to managed secrets (dry-run by default, --apply to commit).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add scripts/backup-db.sh for backing up the embedded PostgreSQL database
with support for custom (.dump) and plain SQL formats, auto-detection of
the configured port, and pruning of backups older than 30 days.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Agent management: hire endpoint with permission gates and pending_approval status,
config revision tracking with rollback, agent duplicate route, permission CRUD.
Block pending_approval agents from auth, heartbeat, and assignments.
Approvals: revision request/resubmit flow, approval comments CRUD, issue-approval
linking, auto-wake agents on approval decisions with context snapshot.
Costs: per-agent breakdown, period filtering (month/week/day/all), cost by agent
list endpoint.
Adapters: agentConfigurationDoc on all adapters, /llms/agent-configuration.txt
reflection routes. Inject PAPERCLIP_APPROVAL_ID, PAPERCLIP_APPROVAL_STATUS,
PAPERCLIP_LINKED_ISSUE_IDS into adapter environments.
Sidebar badges endpoint for pending approval/inbox counts. Dashboard and company
settings extensions.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Rework config store with better file handling. Expand heartbeat-run
command with richer output and error reporting. Improve configure
and onboard commands. Update doctor checks.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Switch from PGlite (WebAssembly) to embedded-postgres for zero-config
local development — provides a real PostgreSQL server with full
compatibility. Add startup banner with config summary on server boot.
Improve server bootstrap with auto port detection, database creation,
and migration on startup. Update DATABASE.md, DEVELOPING.md, and
SPEC-implementation.md to reflect the change. Update CLI database
check and prompts. Simplify OnboardingWizard database options.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add cli/ package with initial scaffolding. Add config-schema to shared
package for typed configuration. Add server config-file loader for
paperclip.config.ts support. Register cli in pnpm workspace. Add
.paperclip/ and .pnpm-store/ to gitignore. Minor Companies page fix.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
New routes: companies, approvals, costs, dashboard, authz. New
services: companies, approvals, costs, dashboard, heartbeat,
activity-log. Add auth middleware and structured error handling.
Expand existing agent and issue routes with richer CRUD operations.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
pnpm monorepo with shared packages, server, and UI workspaces.
Includes Docker multi-stage build, Postgres compose, and root TypeScript/Vitest config.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
