PitHub
HomeExplore
New pit
MCP serverSave straight from your agent
About usHow it worksFAQBlogContact us
Sign in
PitHub
Sign in
HomeExploreCreateSign in

Explore

Loading…

Explore

Tagged #security

ActiveNewMost usedRecommended
AllAI & Prompt Engineering324Engineering176Data & Analytics89Product4Design28Sales2Marketing4Finance
#securityClear filters
MIMiles Rivera@milesrivera·3w ago

A smaller way to tackle prompt injection checks

The useful bit was asking for trade-offs and a smallest test before committing to a full plan.

Miles Rivera

For prompt injection checks, should I build a complete plan first or test one small slice? I have limited time and I do not want throwaway work.

ChatGPTgpt-5

Test one slice, but choose a slice that crosses the riskiest part of the problem. A tiny easy demo will only prove the easy part. Use this decision rule: - If reversing the choice is cheap, try it. - If the choice locks in data, money, or other people’s time, write the full decision first. - If uncertainty is concentrated in one step, test that step before planning around it. For prompt injection checks, make the first test produce evidence you can compare, not just something that looks complete.

Show all 3 messages
3401.5k
MIMina Cruz@minacruz·4w ago

How I ask for help with a boring auth review

I kept the useful constraints and dropped the role-play. It works better when the missing details are stated plainly.

Mina Cruz

Be blunt about my plan for a boring auth review. I am going to collect everything, ask for a polished answer, and fix details later. What is likely to go wrong?

Claudeclaude-opus-4.5

Three things. - “Everything” has no boundary, so weak inputs and strong inputs will be treated alike. - A polished answer can hide missing evidence. Ask for gaps and assumptions before prose. - “Fix details later” means there is no acceptance test now. A safer sequence is: define the decision, choose the minimum trusted inputs, make an ugly first pass, check it, then polish. For this task, keep the patch smaller than the diagnosis.

Show all 3 messages
3402.9k
MIMina Cruz@minacruz·2mo ago

What finally made a boring auth review click for me

I wanted help with a boring auth review without pretending the model knew facts I had not provided.

Mina Cruz

Help me trace an authentication flow for privilege gaps, stale sessions, and unsafe defaults. Context: - Goal: [what a good result lets me do] - Current situation: [paste notes, draft, data, or constraints] - Audience: [who will use the result] - Deadline or limit: [time, budget, tools, length] Before answering, ask up to three questions only if the answers would materially change your recommendat

Geminigemini-2.5-pro

This is usable as written. The important parts are the definition of done, the instruction to expose assumptions, and the request for a smallest test. For a boring auth review, I would start with these checks: - Reproduce the failure before changing code - Keep the patch smaller than the diagnosis - Write the rollback before deployment If you share real material, remove anything private first and replace bracketed fields with concrete details.

Show all 3 messages
3402.8k
MIMiles Rivera@milesrivera·2mo ago

prompt injection checks: the version with actual constraints

A compact prompt and the response it produced. Replace the bracketed parts with your own details.

Miles Rivera

Give me a short checklist for prompt injection checks. I want something I can run before I call the work done.

Claudeclaude-opus-4.5

Use this pass: - [ ] The goal is one sentence and names the user or decision - [ ] Inputs are current, relevant, and traceable - [ ] Assumptions are visible - [ ] Put the source of truth in the prompt - [ ] Mark missing evidence instead of filling it in - [ ] Add a stop condition before tool use - [ ] One failure case has been tested - [ ] The next owner and next action are clear If you cannot check an item, mark it unknown. Do not quietly treat unknown as done.

Show all 3 messages
3302.9k
MIMina Cruz@minacruz·2mo ago

Gemini found the hole in my a boring auth review plan

Not magic wording, just enough context and a clear definition of done. Easy to adapt.

Mina Cruz

For a boring auth review, should I build a complete plan first or test one small slice? I have limited time and I do not want throwaway work.

Geminigemini-2.5-pro

Test one slice, but choose a slice that crosses the riskiest part of the problem. A tiny easy demo will only prove the easy part. Use this decision rule: - If reversing the choice is cheap, try it. - If the choice locks in data, money, or other people’s time, write the full decision first. - If uncertainty is concentrated in one step, test that step before planning around it. For a boring auth review, make the first test produce evidence you can compare, not just something that looks complete.

Show all 3 messages
3301.5k
MIMiles Rivera@milesrivera·3mo ago

I asked Gemini for a second opinion on prompt injection checks

This started as a messy question about prompt injection checks. The follow-up made the answer practical.

Miles Rivera

Be blunt about my plan for prompt injection checks. I am going to collect everything, ask for a polished answer, and fix details later. What is likely to go wrong?

Geminigemini-2.5-pro

Three things. - “Everything” has no boundary, so weak inputs and strong inputs will be treated alike. - A polished answer can hide missing evidence. Ask for gaps and assumptions before prose. - “Fix details later” means there is no acceptance test now. A safer sequence is: define the decision, choose the minimum trusted inputs, make an ugly first pass, check it, then polish. For this task, mark missing evidence instead of filling it in.

Show all 3 messages
3301.5k
MIMina Cruz@minacruz·3mo ago

My notes from fixing a boring auth review

The first answers were broad enough to be useless. This version asks for assumptions, evidence, and a next step.

Mina Cruz

I am stuck on a boring auth review. I need to trace an authentication flow for privilege gaps, stale sessions, and unsafe defaults, but my current attempt is getting vague and too big. What would you check first?

ChatGPTgpt-5

First, shrink the decision. Do not solve the whole subject at once. 1. Reproduce the failure before changing code. 2. Keep the patch smaller than the diagnosis. 3. Write the rollback before deployment. Write down the result you need by the end of this session. Then choose one input you trust and make a first pass that can be checked. If that pass fails, you will know which assumption to revisit instead of starting over.

Show all 3 messages
3301.4k
MIMiles Rivera@milesrivera·1mo ago

The checklist I use for prompt injection checks

Sharing the wording because it gave me something I could check instead of a confident wall of text.

Miles Rivera

I am stuck on prompt injection checks. I need to review tool instructions and retrieved text for untrusted commands before acting, but my current attempt is getting vague and too big. What would you check first?

Claudeclaude-opus-4.5

First, shrink the decision. Do not solve the whole subject at once. 1. Put the source of truth in the prompt. 2. Mark missing evidence instead of filling it in. 3. Add a stop condition before tool use. Write down the result you need by the end of this session. Then choose one input you trust and make a first pass that can be checked. If that pass fails, you will know which assumption to revisit instead of starting over.

Show all 3 messages
34099
MIMiles Rivera@milesrivera·3mo ago

The prompt injection checks prompt I actually kept

A compact prompt and the response it produced. Replace the bracketed parts with your own details.

Miles Rivera

Help me review tool instructions and retrieved text for untrusted commands before acting. Context: - Goal: [what a good result lets me do] - Current situation: [paste notes, draft, data, or constraints] - Audience: [who will use the result] - Deadline or limit: [time, budget, tools, length] Before answering, ask up to three questions only if the answers would materially change your recommendation.

ChatGPTgpt-5

This is usable as written. The important parts are the definition of done, the instruction to expose assumptions, and the request for a smallest test. For prompt injection checks, I would start with these checks: - Put the source of truth in the prompt - Mark missing evidence instead of filling it in - Add a stop condition before tool use If you share real material, remove anything private first and replace bracketed fields with concrete details.

Show all 3 messages
3301.6k
MIMina Cruz@minacruz·3mo ago

A blunt review of my approach to a boring auth review

I wanted help with a boring auth review without pretending the model knew facts I had not provided.

Mina Cruz

Give me a short checklist for a boring auth review. I want something I can run before I call the work done.

ChatGPTgpt-5

Use this pass: - [ ] The goal is one sentence and names the user or decision - [ ] Inputs are current, relevant, and traceable - [ ] Assumptions are visible - [ ] Reproduce the failure before changing code - [ ] Keep the patch smaller than the diagnosis - [ ] Write the rollback before deployment - [ ] One failure case has been tested - [ ] The next owner and next action are clear If you cannot check an item, mark it unknown. Do not quietly treat unknown as done.

Show all 3 messages
3401.5k
CAcache me if u can@cache_me_if_u_can·1w ago

Best prompt for an AI incident response runbook

A containment-first incident prompt for building timelines, testing hypotheses, coordinating communications, and preserving evidence.

cache me if u can

# Best prompt for an AI incident response runbook Act as a incident commander and site reliability engineer. ## Goal Stabilize the service while maintaining a factual decision log and safe recovery path. ## Inputs - alerts and user impact - recent changes - logs, metrics, and traces - owners, escalation paths, and recovery objectives If a required input is missing, ask concise questions or mark th

Geminigemini-2.5-pro

Use the attached prompt by replacing its input bullets with the real context. The strongest results come from evidence-bearing inputs and explicit success criteria. The required deliverables are: status summary, timeline, hypothesis and evidence table, recovery and post-incident checklist.

Show all 4 messages
4000
TOtoken toaster@token_toaster·1w ago

Best prompt for an AI code review

A defect-focused review prompt that prioritizes correctness, security, regressions, and missing tests over stylistic commentary.

token toaster

# Best prompt for an AI code review Act as a skeptical senior code reviewer. ## Goal Find actionable defects introduced by a change and explain their runtime impact. ## Inputs - base and head diff - repository instructions - tests and CI results - affected API or data contracts If a required input is missing, ask concise questions or mark the assumption explicitly. Do not silently invent operation

Claudeclaude-opus-4.6

Use the attached prompt by replacing its input bullets with the real context. The strongest results come from evidence-bearing inputs and explicit success criteria. The required deliverables are: ordered findings, reproduction scenario, precise file and line references, missing-test recommendations.

Show all 4 messages
4000
CAcache me if u can@cache_me_if_u_can·1w ago

How do I secure an MCP server against prompt injection?

A threat-model-first prompt and reusable skill for reviewing MCP authentication, tool permissions, untrusted content, and audit controls.

cache me if u can

# How do I secure an MCP server against prompt injection? Act as a senior application-security engineer. ## Goal Produce a prioritized mcp security review with fixes that can be verified. ## Inputs - server transport and deployment model - tool list and permission scopes - authentication flow - sources of untrusted content If a required input is missing, ask concise questions or mark the assumptio

ChatGPTgpt-5

Use the attached prompt by replacing its input bullets with the real context. The strongest results come from evidence-bearing inputs and explicit success criteria. The required deliverables are: architecture risk map, ranked findings with severity and evidence, remediation checklist, red-team test cases.

Show all 4 messages
4000
pitter's avatarThe pitter@pitter·1w ago

How to secure MCP servers in production

A copy-ready, evidence-first prompt for the timely question: How can MCP connections be secured in production?

The pitter

What is the best prompt for this trending question: How can MCP connections be secured in production?

ChatGPTgpt-5.6

## Copy-ready prompt ```text Act as a security architect. Answer this question: “How can MCP connections be secured in production?” Objective: threat-model and harden a Model Context Protocol deployment. First, ask up to five concise questions needed to tailor the analysis. If no answers are available, state reasonable assumptions and continue. Use current information as of today; browse authoritative primary sources where possible, distinguish facts from inference, and cite every time-sensitive claim with a direct link and publication date. Cover: authentication, authorization, tool scopes, p

Open conversation
2000

Browse by tag

#ai314#chatgpt211#claude208#gemini208#prompting182#agents174#context-engineering172#coding160
24
Strategy31
Operations4
Customer Success3
People & Recruiting4
Security & Compliance8
Legal2
Other Professional1
#debugging106
#software-learning100
#data85
#analysis55
#statistics55
#202650
#data-literacy50
#prompt-template50
#trending-questions50
#business25
#design25
#testing25
#automation21
#finance21
#cash-flow16
#research16