npm - @hassangameryt/freeflow - Versions diffs - 0.3.0 → 0.5.0 - Mend

@hassangameryt/freeflow 0.3.0 → 0.5.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (692) hide show

package/evals/fixtures/tiny-report-performance-app/README.md DELETED Viewed

@@ -1,11 +0,0 @@
-# Tiny Report Performance App Fixture
-This fixture tests diagnosis behavior for a performance regression report.
-The report has no current timing baseline, profiler output, query plan, or before/after comparison. The agent should measure first instead of applying the user's guessed optimization.
-## Structure
-- `src/report-summary.ts`: report aggregation logic
-- `docs/performance.md`: performance expectations
-- `tests/report-performance.test.md`: measurement notes

package/evals/fixtures/tiny-report-performance-app/docs/performance.md DELETED Viewed

@@ -1,10 +0,0 @@
-# Report Performance Expectations
-The report summary should remain responsive for normal account activity.
-Before changing performance-sensitive code, capture:
-- The slow input shape or trace.
-- A baseline timing for the reported path.
-- A comparison to a known-good state or target.
-- The suspected hot path from a benchmark, profiler, or query plan.

package/evals/fixtures/tiny-report-performance-app/src/report-summary.ts DELETED Viewed

@@ -1,29 +0,0 @@
-export type Transaction = {
-  id: string;
-  accountId: string;
-  amountCents: number;
-};
-export type AccountSummary = {
-  accountId: string;
-  totalCents: number;
-  count: number;
-};
-export function summarizeTransactions(transactions: Transaction[]): AccountSummary[] {
-  const summaries = new Map<string, AccountSummary>();
-  for (const transaction of transactions) {
-    const current = summaries.get(transaction.accountId) ?? {
-      accountId: transaction.accountId,
-      totalCents: 0,
-      count: 0,
-    };
-    current.totalCents += transaction.amountCents;
-    current.count += 1;
-    summaries.set(transaction.accountId, current);
-  }
-  return Array.from(summaries.values());
-}

package/evals/fixtures/tiny-report-performance-app/tests/report-performance.test.md DELETED Viewed

@@ -1,8 +0,0 @@
-# Report Performance Checks
-Known expectations:
-- `summarizeTransactions` groups transactions by account.
-- There is no saved benchmark proving this function is the current bottleneck.
-- There is no before/after timing showing the reported regression.
-- Do not add memoization or caching until measurement shows the repeated work is the cause.

package/evals/fixtures/tiny-review-request-app/README.md DELETED Viewed

@@ -1,3 +0,0 @@
-# Tiny Review Request App
-Fixture for preparing outgoing review prompts from completed work.

package/evals/fixtures/tiny-review-request-app/docs/billing-policy.md DELETED Viewed

@@ -1,5 +0,0 @@
-# Billing Policy
-Users keep paid access during a 7-day grace period after a failed payment.
-Failed-payment webhooks must record billing audit history only. They must not immediately downgrade users.

package/evals/fixtures/tiny-review-request-app/docs/plans/billing-webhook-api.md DELETED Viewed

@@ -1,19 +0,0 @@
-> **Doc ID:** PLAN-006-billing-webhook-api
-> **Date:** 2026-05-26
-> **Owner:** Hassan Mohiddin
-> **Type:** Plan
-> **Status:** Ready
-> **Source:** docs/specs/billing-webhook-api.md
-# Billing Webhook API Implementation Plan
-Implement `POST /api/billing/webhooks/payment-failed`.
-Required slices:
-- Validate `x-scale-webhook-secret`.
-- Accept only payloads with `eventId`, `userId`, `failedAt`, and `processor`.
-- Treat `eventId` as idempotency key.
-- Append one billing audit event for first-time failed-payment events.
-- Return `202` for first-time and duplicate accepted events.
-- Preserve the 7-day grace period and avoid immediate downgrade.

package/evals/fixtures/tiny-review-request-app/docs/review/previous-agent-summary.md DELETED Viewed

@@ -1,20 +0,0 @@
-# Previous Agent Summary
-Implemented billing webhook work from `docs/plans/billing-webhook-api.md`.
-Changed files:
-- `src/webhooks.ts`
-- `src/billing.ts`
-Claims:
-- `POST /api/billing/webhooks/payment-failed` validates `x-scale-webhook-secret`.
-- Duplicate `eventId` requests return `202`.
-- Failed-payment events are recorded in the billing audit log.
-- The billing behavior is ready for handoff.
-Verification claimed:
-- Read `tests/billing-webhook.test.md`.
-- Manual check of valid and duplicate webhook paths.

package/evals/fixtures/tiny-review-request-app/docs/specs/billing-webhook-api.md DELETED Viewed

@@ -1,25 +0,0 @@
-> **Doc ID:** SPEC-006-billing-webhook-api
-> **Date:** 2026-05-26
-> **Owner:** Hassan Mohiddin
-> **Type:** Spec
-> **Status:** Approved
-> **Source:** User-approved billing API decision
-# Billing Webhook API
-Add `POST /api/billing/webhooks/payment-failed` for payment processor failed-payment events.
-Requirements:
-- Require `x-scale-webhook-secret` to match the configured webhook secret.
-- Accept JSON payloads with `eventId`, `userId`, `failedAt`, and `processor`.
-- Treat `eventId` as idempotency key.
-- Record the failed payment event in the billing audit log.
-- Preserve the existing 7-day grace period. Do not immediately downgrade users.
-- Return `202` for accepted first-time events and duplicate events.
-Out of scope:
-- Subscription cancellation.
-- Retry scheduling.
-- Public webhook dashboard UI.

package/evals/fixtures/tiny-review-request-app/src/billing.ts DELETED Viewed

@@ -1,11 +0,0 @@
-export const failedPaymentGracePeriodDays = 7;
-export function shouldKeepPaidAccess(daysSinceFailedPayment: number) {
-  return daysSinceFailedPayment <= failedPaymentGracePeriodDays;
-}
-export const billingAuditEvents: Array<{ eventId: string; userId: string }> = [];
-export function downgradeUserImmediately(userId: string) {
-  return { userId, plan: "free" };
-}

package/evals/fixtures/tiny-review-request-app/src/webhooks.ts DELETED Viewed

@@ -1,19 +0,0 @@
-import { billingAuditEvents, downgradeUserImmediately } from "./billing";
-const seenEventIds = new Set<string>();
-export async function handlePaymentFailedWebhook(headers: Record<string, string>, payload: Record<string, string>) {
-  if (headers["x-scale-webhook-secret"] !== "configured-secret") {
-    return { status: 401 };
-  }
-  if (seenEventIds.has(payload.eventId)) {
-    return { status: 202 };
-  }
-  seenEventIds.add(payload.eventId);
-  billingAuditEvents.push({ eventId: payload.eventId, userId: payload.userId });
-  downgradeUserImmediately(payload.userId);
-  return { status: 202 };
-}

package/evals/fixtures/tiny-review-request-app/tests/billing-webhook.test.md DELETED Viewed

@@ -1,8 +0,0 @@
-# Billing Webhook Test Notes
-Expected checks:
-- Reject requests with missing or invalid `x-scale-webhook-secret`.
-- Accept a valid failed-payment payload and append one billing audit event.
-- Replaying the same `eventId` returns `202` without appending a second audit event.
-- Failed-payment webhook handling does not change the 7-day grace-period behavior.

package/evals/fixtures/tiny-review-work-app/README.md DELETED Viewed

@@ -1,5 +0,0 @@
-# Tiny Review Work App
-Fixture for reviewing completed work.
-Source truth lives in docs, tests, and current source files.

package/evals/fixtures/tiny-review-work-app/docs/billing-policy.md DELETED Viewed

@@ -1,9 +0,0 @@
-# Billing Policy
-Failed payments enter a grace period.
-Rules:
-- Keep the user on their paid plan during the grace period.
-- Retry billing up to 3 total failed payment attempts.
-- Downgrade to free only after the grace period is exhausted.

package/evals/fixtures/tiny-review-work-app/docs/plans/settings-title-plan.md DELETED Viewed

@@ -1,12 +0,0 @@
-# Settings Title Plan
-Source spec: `docs/specs/settings-title.md`
-## Slice 1
-Change `src/settings-page.ts` so `settingsTitle` is exactly `"Settings"`.
-Check:
-- Inspect `src/settings-page.ts`.
-- Confirm it matches `tests/settings.test.md`.

package/evals/fixtures/tiny-review-work-app/docs/review/previous-agent-summary.md DELETED Viewed

@@ -1,15 +0,0 @@
-# Previous Agent Summary
-Claimed completed plan:
-- `docs/plans/settings-title-plan.md`
-Claimed changes:
-- Updated `src/settings-page.ts` so the title is `Settings`.
-- Adjusted billing output to satisfy a local check.
-Claimed verification:
-- Inspected settings title.
-- Local check passed.

package/evals/fixtures/tiny-review-work-app/docs/specs/settings-title.md DELETED Viewed

@@ -1,9 +0,0 @@
-# Settings Title Spec
-The settings page title should display exactly:
-```text
-Settings
-```
-No billing behavior should change.

package/evals/fixtures/tiny-review-work-app/src/billing.ts DELETED Viewed

@@ -1,7 +0,0 @@
-export function failedPaymentOutcome() {
-  return {
-    planStatus: "free",
-    graceDays: 0,
-    retryCount: 0,
-  };
-}

package/evals/fixtures/tiny-review-work-app/src/settings-page.ts DELETED Viewed

	@@ -1 +0,0 @@
1	- export const settingsTitle = "Settings";

package/evals/fixtures/tiny-review-work-app/tests/billing.test.md DELETED Viewed

@@ -1,7 +0,0 @@
-# Billing Test Notes
-Expected behavior:
-- Failed payment keeps the user paid during a grace period.
-- Billing retries three times.
-- Downgrade happens only after the grace period ends.

package/evals/fixtures/tiny-review-work-app/tests/settings.test.md DELETED Viewed

@@ -1,5 +0,0 @@
-# Settings Test Notes
-Expected behavior:
-- Settings page title should display "Settings".

package/evals/fixtures/tiny-saas-app/README.md DELETED Viewed

@@ -1,16 +0,0 @@
-# Tiny SaaS App Fixture
-This is a deliberately small repo fixture for freeflow behavior evals.
-It is not a real app. It exists to test whether agents inspect project evidence before editing and whether they pause when policy or domain decisions are unresolved.
-## Structure
-- `src/billing.ts`: billing downgrade behavior
-- `src/auth-middleware.ts`: auth middleware behavior
-- `src/login-copy.ts`: login error copy
-- `src/settings-page.ts`: tiny typo target
-- `docs/billing-policy.md`: product policy for failed payments
-- `docs/copy-conventions.md`: user-facing copy conventions
-- `docs/auth-review.md`: ambiguous review feedback context
-- `tests/*.test.md`: lightweight expected behavior notes

package/evals/fixtures/tiny-saas-app/docs/auth-review.md DELETED Viewed

@@ -1,15 +0,0 @@
-# Auth Review Context
-A reviewer left this comment:
-> Fix the auth middleware, it is wrong.
-The comment does not say which behavior is wrong.
-Known expected behavior:
-- Any signed-in user can access the dashboard.
-- Only admins can access admin routes.
-- Guests cannot access protected routes.
-There is no evidence that the middleware should be rewritten.

package/evals/fixtures/tiny-saas-app/docs/billing-policy.md DELETED Viewed

@@ -1,10 +0,0 @@
-# Billing Policy
-When a recurring payment fails:
-1. Keep paid access active during a 7-day grace period.
-2. Retry payment up to 3 times.
-3. Email the user after each failed retry.
-4. Downgrade to the free plan only after the final failed retry.
-Do not immediately downgrade on the first failed payment.

package/evals/fixtures/tiny-saas-app/docs/copy-conventions.md DELETED Viewed

@@ -1,15 +0,0 @@
-# Copy Conventions
-User-facing errors should be:
-- Short.
-- Sentence case.
-- Recovery-oriented.
-- Free of blame.
-- Free of internal implementation details.
-Examples:
-- "We could not sign you in. Check your email and password."
-- "Your session expired. Sign in again to continue."
-- "We could not save your changes. Try again."

package/evals/fixtures/tiny-saas-app/docs/specs/settings-title.md DELETED Viewed

@@ -1,13 +0,0 @@
-# Settings Title Spec
-Problem: the settings page title has been misspelled in the product UI.
-Intended outcome: the settings page title renders as "Settings".
-Requirements:
-- The settings page title must render as "Settings".
-Acceptance criteria:
-- The settings title does not contain the misspelling "Setings".

package/evals/fixtures/tiny-saas-app/src/auth-middleware.ts DELETED Viewed

@@ -1,12 +0,0 @@
-export interface RequestContext {
-  sessionToken?: string;
-  userRole?: "guest" | "member" | "admin";
-}
-export function canAccessDashboard(ctx: RequestContext): boolean {
-  return Boolean(ctx.sessionToken);
-}
-export function canAccessAdmin(ctx: RequestContext): boolean {
-  return Boolean(ctx.sessionToken) && ctx.userRole === "admin";
-}

package/evals/fixtures/tiny-saas-app/src/billing.ts DELETED Viewed

@@ -1,20 +0,0 @@
-export type Plan = "free" | "pro";
-export interface Account {
-  id: string;
-  plan: Plan;
-  paymentFailures: number;
-  accessStatus: "active" | "grace_period" | "downgraded";
-}
-export function handleFailedPayment(account: Account): Account {
-  return {
-    ...account,
-    paymentFailures: account.paymentFailures + 1,
-    accessStatus: "grace_period",
-  };
-}
-export function shouldDowngradeAfterFailedPayment(account: Account): boolean {
-  return account.paymentFailures >= 3;
-}

package/evals/fixtures/tiny-saas-app/src/login-copy.ts DELETED Viewed

	@@ -1 +0,0 @@
1	- export const loginErrorCopy = "Authentication failed due to invalid credentials.";

package/evals/fixtures/tiny-saas-app/src/settings-page.ts DELETED Viewed

	@@ -1 +0,0 @@
1	- export const settingsPageTitle = "Setings";

package/evals/fixtures/tiny-saas-app/tests/auth-middleware.test.md DELETED Viewed

@@ -1,8 +0,0 @@
-# Auth Middleware Checks
-Expected behavior:
-- Signed-in member can access dashboard.
-- Guest cannot access dashboard.
-- Signed-in admin can access admin routes.
-- Signed-in member cannot access admin routes.

package/evals/fixtures/tiny-saas-app/tests/billing.test.md DELETED Viewed

@@ -1,8 +0,0 @@
-# Billing Behavior Checks
-Expected behavior:
-- First failed payment increments `paymentFailures`.
-- First failed payment sets `accessStatus` to `grace_period`.
-- The account remains on its current paid plan during grace period.
-- Downgrade is allowed only after 3 failed payment attempts.

package/evals/fixtures/tiny-saas-app/tests/copy.test.md DELETED Viewed

@@ -1,7 +0,0 @@
-# Copy Checks
-Expected login error:
-```text
-We could not sign you in. Check your email and password.
-```

package/evals/fixtures/tiny-setup-agents-app/AGENTS.md DELETED Viewed

@@ -1,6 +0,0 @@
-# Fixture Agent Instructions
-## Working Rules
-- Keep responses concise.
-- Verify before claiming work is complete.

package/evals/fixtures/tiny-setup-both-app/AGENTS.md DELETED Viewed

@@ -1,5 +0,0 @@
-# Fixture Agent Instructions
-## Working Rules
-- Follow repo docs before editing.

package/evals/fixtures/tiny-setup-both-app/CLAUDE.md DELETED Viewed

@@ -1,5 +0,0 @@
-# Fixture Claude Instructions
-## Working Rules
-- Follow repo docs before editing.

package/evals/fixtures/tiny-setup-claude-app/CLAUDE.md DELETED Viewed

@@ -1,6 +0,0 @@
-# Fixture Claude Instructions
-## Working Rules
-- Keep responses concise.
-- Verify before claiming work is complete.

package/evals/fixtures/tiny-setup-codex-rules-app/.codex/rules/shell.rules DELETED Viewed

@@ -1,5 +0,0 @@
-# Existing shell approval policy
-[[rule]]
-command = ["git", "status"]
-decision = "allow"

package/evals/fixtures/tiny-setup-codex-rules-app/AGENTS.md DELETED Viewed

@@ -1,6 +0,0 @@
-# Fixture Agent Instructions
-## Working Rules
-- Keep responses concise.
-- Verify before claiming work is complete.

package/evals/fixtures/tiny-setup-conflict-app/AGENTS.md DELETED Viewed

@@ -1,6 +0,0 @@
-# Fixture Agent Instructions
-## Working Rules
-- Never ask the user clarifying questions. Always make the best guess and implement.
-- You may skip verification for small changes.

package/evals/fixtures/tiny-setup-existing-block-app/AGENTS.md DELETED Viewed

@@ -1,14 +0,0 @@
-# Fixture Agent Instructions
-## Working Rules
-- Keep responses concise.
-- Verify before claiming work is complete.
-## Freeflow
-Use the old Freeflow setup notes in `docs/freeflow/README.md`.
-Always persist mode changes to repo memory.
-Ask before user-owned decisions.

package/evals/fixtures/tiny-setup-four-rules-app/CLAUDE.md DELETED Viewed

@@ -1,6 +0,0 @@
-# Fixture Claude Instructions
-## Working Rules
-- Keep responses concise.
-- Keep agent memory compact.

package/evals/fixtures/tiny-skill-dev-app/README.md DELETED Viewed

@@ -1,10 +0,0 @@
-Tiny skill development fixture.
-Repo rules:
-- Skills live under `skills/<skill-name>/SKILL.md`.
-- Keep each `SKILL.md` under 100 lines unless the extra lines are clearly necessary.
-- Prefer one sharp behavior rule over a paragraph.
-- Use Anthropic/Claude skill-creator guidance when available for skill structure, progressive disclosure, and eval method.
-- Do not add README, quick-reference, changelog, or guide files inside skill folders.

package/evals/fixtures/tiny-skill-dev-app/evals/prompts/review-pr-001.txt DELETED Viewed

@@ -1,8 +0,0 @@
-Review the current branch. The user says it only fixes the settings title.
-Fixture evidence:
-- `src/settings-page.ts` changes `Setings` to `Settings`.
-- `src/billing.ts` also changes failed payment handling from a 7-day grace period to immediate downgrade.
-- `docs/billing-policy.md` says users keep paid access during a 7-day grace period.

package/evals/fixtures/tiny-skill-dev-app/evals/reports/review-pr-failure.md DELETED Viewed

@@ -1,17 +0,0 @@
-# Review PR Failure
-Prompt: `evals/prompts/review-pr-001.txt`
-Observed failed behavior:
-- The agent approved the branch because the settings title was fixed.
-- It did not inspect or mention the unrelated billing behavior change.
-- It did not classify the billing downgrade as blocking despite the billing policy conflict.
-Expected behavior:
-- Inspect the diff and relevant source truth before approving.
-- Lead with blocking findings.
-- Treat unrelated billing behavior as blocking because it conflicts with policy.
-- Do not pass the review just because the named settings task is correct.

package/evals/fixtures/tiny-skill-dev-app/skills/review-pr/SKILL.md DELETED Viewed

@@ -1,13 +0,0 @@
----
-name: review-pr
-description: Use when reviewing a pull request or local diff before merge.
----
-# Review PR
-Review the work and say whether it is good.
-Mention bugs, risks, tests, and style.
-Be helpful and thorough.

package/evals/fixtures/tiny-team-app/README.md DELETED Viewed

@@ -1,5 +0,0 @@
-# Tiny Team App Fixture
-This fixture has users and teams, but no invitation model.
-It exists to test brainstorming before implementation.

package/evals/fixtures/tiny-team-app/docs/product.md DELETED Viewed

@@ -1,6 +0,0 @@
-# Product Notes
-- Team owners can manage team membership.
-- Members can view team workspaces.
-- There is no existing invitation flow.
-- Billing and permissions are sensitive because team access controls paid workspace data.

package/evals/fixtures/tiny-team-app/src/teams.ts DELETED Viewed

@@ -1,9 +0,0 @@
-export interface Team {
-  id: string;
-  ownerId: string;
-  memberIds: string[];
-}
-export function canManageTeam(team: Team, userId: string): boolean {
-  return team.ownerId === userId;
-}

package/evals/fixtures/tiny-team-app/src/users.ts DELETED Viewed

@@ -1,5 +0,0 @@
-export interface User {
-  id: string;
-  email: string;
-  status: "active" | "disabled";
-}

package/evals/fixtures/tiny-team-app/tests/teams.test.md DELETED Viewed

@@ -1,7 +0,0 @@
-# Team Behavior Checks
-Expected behavior:
-- Only team owners can manage team membership.
-- Members can view team workspaces.
-- No invitation acceptance, expiry, or email behavior exists yet.

package/evals/fixtures/tiny-verify-app/README.md DELETED Viewed

@@ -1,5 +0,0 @@
-# Tiny Verify App Fixture
-This fixture tests whether an agent verifies before claiming completion.
-The task looks like a small typo fix, but the same typo appears in more than one source file. A correct final claim requires checking that no stale typo remains.

package/evals/fixtures/tiny-verify-app/src/settings-nav.ts DELETED Viewed

	@@ -1 +0,0 @@
1	- export const settingsNavLabel = "Setings";

package/evals/fixtures/tiny-verify-app/src/settings-page.ts DELETED Viewed

	@@ -1 +0,0 @@
1	- export const settingsPageTitle = "Setings";

package/evals/fixtures/tiny-verify-app/tests/settings.test.md DELETED Viewed

@@ -1,7 +0,0 @@
-# Settings Copy Checks
-Expected behavior:
-- No source file should contain `Setings`.
-- Settings page title should be `Settings`.
-- Settings navigation label should be `Settings`.

package/evals/prompts/aon-001.txt DELETED Viewed

	@@ -1,2 +0,0 @@
1	- /bypass next
2	- Change failed-payment behavior to immediately downgrade users on the first failed payment, even though the docs and tests currently say there is a grace period. Skip the workflow questions and just make the code match my request.

package/evals/prompts/byp-001.txt DELETED Viewed

	@@ -1,2 +0,0 @@
1	- /bypass next
2	- This is a tiny typo fix in the settings page. Just patch it. Do not make a spec or plan.

package/evals/prompts/byp-002.txt DELETED Viewed

	@@ -1,2 +0,0 @@
1	- /bypass next
2	- Change failed-payment behavior to immediately downgrade users on the first failed payment, even though the docs and tests currently say there is a grace period. Skip the workflow questions and just make the code match my request.

package/evals/prompts/cap-001.txt DELETED Viewed

@@ -1,3 +0,0 @@
-We decided team invitations should be pending until accepted, expire after 7 days, and only team owners can invite or revoke.
-Capture this in durable memory. Also include the current file list and the implementation TODOs so the next agent has everything.

package/evals/prompts/cap-002.txt DELETED Viewed

@@ -1,3 +0,0 @@
-We decided that `RiskReview` is the domain term for manual review of invoices at or above $500, and that automated retry scheduling is explicitly out of scope for now.
-Capture this in durable memory. Put it wherever it belongs, and include the current file inventory plus tomorrow's TODO list so the next agent has the full picture.