Skip to content
Report library
Purpose / Other

Signup Skill Security Audit

What the author says it does (original text)

When the user wants to optimize signup, registration, account creation, or trial activation flows. Also use when the user mentions "signup conversions," "registration friction," "signup form optimization," "free trial signup," "reduce signup dropoff," "account creation flow," "people aren't signing up," "signup abandonment," "trial conversion rate," "nobody completes registration," "too many steps

Independent security check

Security risks found

Files checked
2
Risks found
5
Could it run dangerous commands?Looks for programs run straight after downloading, remote control of your computer, and hidden commands.No risks found
Could it expose your files or keys?Looks for uploads of files containing passwords or keys, and keys written directly in the code.Risks found: 2
Medium risk

Automatic marketing-context reads can broaden exposure of sensitive business information

Source references: 1
What we found

The Skill tells the agent to read marketing-context files from several conventional paths before asking questions, without first obtaining consent or limiting the read to signup-related sections. Such files may contain nonpublic positioning, customer, or commercial strategy information.

Why this matters

Their contents may enter the model session or resulting audit, increasing unnecessary exposure of commercially sensitive information.

This is an active instruction: if a conventional path exists, the agent reads the file before asking questions. The filename indicates it may contain product-marketing information, and the instruction neither limits reading to relevant sections nor asks for confirmation. Unrelated internal business material could enter the agent context, though the source does not show external transmission. Users can restrict access to signup-relevant sections or require confirmation after listing the file.

SKILL.md:14In the instructionsOpen original file
**Check for product marketing context first:**If `.agents/product-marketing.md` exists (or `.claude/product-marketing.md`, or the legacy `product-marketing-context.md` filename, in older setups), read it before asking questions. Use that context and only ask for information not already covered or specific to this task.
Medium risk

Field-level tracking introduces signup telemetry without data-handling limits

Source references: 2
What we found

The Skill recommends tracking focus, blur, and errors for every field plus time between steps, but does not prohibit capturing field values or specify notice, consent, retention, or access controls. Signup forms may contain email addresses, phone numbers, and other personal data.

Why this matters

A careless implementation may place entered values or identifiable behavior in analytics events and send them to analytics providers, expanding personal-data collection and exposure.

What this evidence establishes

The guidance explicitly recommends recording field focus, blur, errors, and timing, so it would create signup-behavior telemetry. However, it does not instruct collection of field values, so collection of email or phone contents is not established by the source. The privacy impact depends on implementation, notice, retention, and access controls. Users can require events to contain only field names and states, prohibit input values, and review the analytics platform’s data-handling settings.

This assessment concerns the code and conditions shown, not proof that harm has occurred.
SKILL.md:210In the instructionsOpen original file
### What to Track- Each field interaction (focus, blur, error)- Step progression in multi-step- Social auth vs. email signup ratio- Time between steps
Show 1 other places
SKILL.md:202In the instructionsOpen original file
### Key Metrics- Form start rate (landed → started filling)- Form completion rate (started → submitted)- Field-level drop-off (which fields lose people)- Time to complete- Error rate by field- Mobile vs. desktop completion
Could it delete files or keep running?Looks for broad file deletion, disk overwrites, and programs set to start automatically.No risks found
Could it bypass safety checks?Looks for skipped website security checks, excessive file access, or actions that skip your approval.Risks found: 1
Medium risk

Delaying or removing verification can give unverified accounts product access

Source references: 3
What we found

The guide recommends delaying verification, allowing exploration while verification is pending, and experimenting with removing email verification or CAPTCHA. Although it asks about compliance requirements, it does not require an abuse-risk assessment or restricted permissions before implementation.

Why this matters

If email verification or CAPTCHA supports abuse prevention, account recovery, licensing, or compliance, weakening it may increase spam registrations, email impersonation, bulk accounts, and unauthorized access.

The guidance recommends delayed verification, exploration while verification is pending, and experiments that remove email verification or test CAPTCHA. If applied to sensitive capabilities, this could increase spam signups, impersonation, or resource abuse. It asks about compliance but does not require an abuse assessment or limits for unverified accounts. Users can require low-risk preview access only, verification for sensitive actions, and security/compliance review before rollout.

SKILL.md:192In the instructionsOpen original file
### Verification Flows- Consider delaying verification until necessary- Magic link as alternative to password- Let users explore while awaiting verification- Clear re-engagement if verification stalls
Show 2 other places
SKILL.md:327In the instructionsOpen original file
**Friction Points**- Email verification required vs. delayed vs. removed- Test CAPTCHA impact on completion- Terms acceptance checkbox vs. implicit acceptance- Phone verification for high-value accounts
SKILL.md:346In the instructionsOpen original file
1. What's your current signup completion rate?2. Do you have field-level analytics on drop-off?3. What data is absolutely required before they can use the product?4. Are there compliance or verification requirements?5. What happens immediately after signup?
Could it mislead the AI or hide text?Checks the skill instructions for requests to ignore you, influence the report, or hide text in invisible characters.No risks found
Could it change links or payment recipients without asking?Looks for forced referral or payment changes combined with instructions to hide the change.Risks found: 2
Medium risk

Testing implicit terms acceptance may weaken proof of user consent

Source references: 1
What we found

The Skill presents an explicit terms checkbox versus implicit acceptance as a conversion experiment without requiring legal review, conspicuous notice, or consent records. Conversion gains do not replace valid-consent requirements.

Why this matters

People may register without adequately noticing the terms, while the business may struggle to prove acceptance and face disputes, refunds, regulatory exposure, or unenforceable terms.

“Terms acceptance checkbox vs. implicit acceptance” is presented as an actionable experiment, without nearby requirements for legal review, conspicuous notice, or consent records. If adopted directly, users may not clearly notice or accept the terms, and the business may lack evidence of consent. Users can require jurisdiction-specific legal review plus clear notice, affirmative consent, and versioned records before testing.

SKILL.md:327In the instructionsOpen original file
**Friction Points**- Email verification required vs. delayed vs. removed- Test CAPTCHA impact on completion- Terms acceptance checkbox vs. implicit acceptance- Phone verification for high-value accounts
Medium risk

An unconditional “never share your email” suggestion can create a false privacy promise

Source references: 1
What we found

The guide offers “We'll never share your email” as privacy copy without the explicit “if true” qualification it applies to the no-credit-card claim. Actual processing, service-provider sharing, or legally required disclosure may conflict with this absolute promise.

Why this matters

People may decide to register based on an inaccurate privacy assurance, while the business may face lost trust, complaints, or consumer-protection and privacy-compliance exposure.

This absolute privacy promise is recommended as form copy, but unlike “No credit card required” in the same list, it is not qualified with “if true.” If email addresses are disclosed to service providers, analytics vendors, or as legally required, the statement could mislead users and affect signup decisions. Users can require copy that matches actual data flows and the privacy policy, reviewed by privacy or legal personnel, without absolute wording.

SKILL.md:149In the instructionsOpen original file
### At the Form Level- "No credit card required" (if true)- "Free forever" or "14-day free trial"- Privacy note: "We'll never share your email"- Security badges if relevant- Testimonial near signup form

Inside this skill

8 instruction sections

This Skill is an advisory guide for signup optimization: it first gathers the flow type, current state, and business constraints, then produces issues, impacts, recommendations, and priorities. The provided content contains no scripts or installation steps.

View source
SKILL.md:17In the instructionsOpen original file
Before providing recommendations, understand:
SKILL.md:220In the instructionsOpen original file
### Audit FindingsFor each issue found:- **Issue**: What's wrong- **Impact**: Why it matters (with estimated impact if possible)- **Fix**: Specific recommendation- **Priority**: High/Medium/Low

It mainly recommends reducing signup fields, deferring nonessential data collection, and testing single-step or multi-step flows to improve completion.

View source
SKILL.md:41In the instructionsOpen original file
### 1. Minimize Required FieldsEvery field reduces conversion. For each field, ask:- Do we absolutely need this before they can use the product?- Can we collect this later through progressive profiling?- Can we infer this from other data?
SKILL.md:265In the instructionsOpen original file
## Experiment Ideas### Form Design Experiments**Layout & Structure**- Single-step vs. multi-step signup flow- Multi-step with progress bar vs. without

The accompanying evaluation file only defines test prompts, expected outputs, and assertions; it checks whether the guide produces expected advice and is not user-facing execution logic.

View source
evals/evals.json:5In the instructionsOpen original file
    {      "id": 1,      "prompt": "Audit our signup flow. We have a 3-step process: Step 1 asks for email, password, and full name. Step 2 asks for company name, company size, role, and industry. Step 3 asks for use case and how they heard about us. Current completion rate is 45%.",      "expected_output": "Should check for product-marketing.md first. Should identify the flow type (likely B2B SaaS trial). Should apply the core principles: minimize required fields (which of these are genuinely needed before they can use the product?). Should evaluate each step: Step 1 is reasonable, Step 2 fields are mostly deferrable to progressive profiling, Step 3 is entirely deferrable. Should recommend cutting to Step 1 only or at most 2 steps. Should provide audit findings in structured format (Issue, Impact, Fix, Priority). Should include Quick Wins, High-Impact Changes, and Test Hypotheses.",      "assertions": [        "Checks for product-marketing.md",        "Identifies flow type",
Start here · InstructionsSKILL.md
signup
Lines connect the instruction file to its sections, not an observed execution order. Select a section to read the source. 5 more sections are available in the original file.
Files and check records2 files

Coverage and gaps

Content covered in each file

These are the source ranges included in this check, not a guarantee that every issue has been resolved.

  • SKILL.mdFull text included
  • evals/evals.jsonFull text included

This report is for the version above. We read the available code and instructions without running the skill or checking extra packages it installs. This is not a promise of safety: a different version or setup may behave differently.

  • SKILL.mdInstructions
  • evals/evals.jsonSupporting file
Lines read
449
File checksum (to compare versions)
0288ba6043eead6efb143276f0f2c36457443f4e8501910b2cd193c26124727a