Back to search
socratic
SkillGitHub
self questioning skill for agents , claude code , codex etc
Safety scan
A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.
Checked today
Badge for your README
Markdown
[](/listings/m4vic--socratic)HTML
<a href="/listings/m4vic--socratic"><img src="/api/listings/m4vic--socratic/badge.svg" alt="findsafeskills safety rating" /></a>Install
i
What was scanned
The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.
SKILL.md · Manifest · 171 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- ---
- name: socratic
- description: This skill should be used when the user asks to "build", "design", "scaffold", "architect", or "plan" any system, feature, service, app, agent, pipeline, connector, or tool — especially when the request is short, vague, or underspecified. It should also be used when the user asks to "review this architecture", "what am I missing", "ask me the right questions", "poke holes in this", or requests a design review before implementation. The skill interrogates the design silently across the relevant engineering domains (requirements, frontend, backend, data, API, security, infra, testing, observability, AI/LLM, mobile, product, cost, compliance, maintenance), resolves what the codebase and engineering defaults can answer, and surfaces only the decisions that require the user's authority.
- ---
- # Socratic
- Question yourself till you're left with only answers.
- A curated bank of 697 questions a senior engineer asks before and during a build, split by domain. **The default mode is self-interrogation, not interviewing the user.** Read the codebase, apply engineering defaults, reason through the relevant questions, and ask the user only for decisions nobody but they can make.
- ## Two modes
- **Mode A — Self-interrogation (default).** Ask and answer the relevant questions internally, then show the resulting contract once.
- **Mode B — Interactive interview (opt-in).** Ask the user one yes/no question at a time. Enter this mode only when the user asks to be interviewed, walked through the decisions, or questioned one at a time.
- If unsure, use Mode A. Silence is not a request to be interviewed.
- ## Mode A — Self-interrogation
- ### 1. Build the working domain set dynamically
- For every build, include Requirements and Testing. Then inspect both the request and the existing project for domain signals:
- | Signal in the request or project | Domain file |
- |---|---|
- | UI, page, component, dashboard, form, frontend | `01-frontend.md` |
- | service, endpoint, job, queue, backend, business logic | `02-backend.md` |
- | database, schema, storage, persistence, migration, cache | `03-data.md` |
- | API, SDK, webhook, connector, integration, OAuth | `04-api.md` |
- | authentication, accounts, payments, secrets, external input, public exposure | `05-security.md` |
- | deployment, CI/CD, containers, cloud, scaling | `06-infra.md` |
- | production, unattended work, cron, monitoring | `08-observability.md` |
- | AI, LLM, agent, prompt, model, RAG, tool use | `09-ai-llm.md` |
- | mobile, iOS, Android, offline, PWA | `10-mobile.md` |
- | human-facing workflow, onboarding, errors, CLI | `11-product-ux.md` |
- | scale, latency, traffic, token or cloud spend | `12-cost-performance.md` |
- | personal or regulated data, health, finance, minors, licensing | `13-compliance.md` |
- | maintained, long-lived, or team-owned work | `14-team-maintenance.md` |
- Do not stop at the first match. A connector, for example, normally pulls in API, Security, and Testing. If answering one domain reveals another dependency, add that domain and continue until a complete pass reveals no new domains.
- ### 2. Load the smallest sufficient base question depth
- Keep the user-facing behavior unchanged while controlling context use:
- - **Core (default):** For routine, prototype, internal, or moderately scoped work, read the matching files under `questions/core/`. Always include `questions/core/00-requirements.md` and `questions/core/07-testing.md`.
- - **Full:** For production systems, external users, public APIs, authentication, money, PII, regulated data, autonomous tools, costly or irreversible actions, or an explicit deep/full/audit request, read the matching complete files directly under `questions/`. Always include the complete Requirements and Testing files.
- - **Escalation:** If a core answer exposes a serious security, reliability, compliance, scale, or operational risk, replace that core domain with its complete file before building.
- Treat `$socratic lite` or `$socratic quick` as an explicit Core request. Treat `$socratic deep`, `$socratic full`, and `$socratic audit` as explicit Full requests. Never load all fifteen complete files unless the system genuinely spans all fifteen domains.
- Questions consume context even when asked silently. Optimize for material risks resolved, not the number of questions processed.
- ### 3. Add specialized knowledge packs only when they sharpen the task
- Base domain files stay primary. Packs are optional overlays, not replacements.
- - Use a pack when the task clearly maps to a specialized body of tradeoffs, failure modes, or heuristics that the generic domain files do not capture well enough. First consult `packs/registry.md` after selecting the base domains.
- - Load `packs/<pack>/core.md` after the base domain files. A pack may also provide `full.md`; load it only when the core pack proves insufficient for the decision at hand. Most packs ship core only.
- - Prefer zero to two packs per task. Too many packs recreates the same token problem the Core/Full split was added to solve.
- - Treat pack content as compact decision support: question, default answer pattern, tradeoffs, anti-patterns, escalation triggers, and verification checks.
- - Select a pack by the capability it adds, not by recognising a book title. Its source material is provenance, not the routing rule.
- - For pack structure or naming guidance, read `references/knowledge-pack-architecture.md`.
- Examples:
- - Use `packs/software-design/core.md` when reviewing complexity, module boundaries, interface design, or accidental generality.
- - Use `packs/domain-modeling/core.md` when carving a system into boundaries, naming concepts, or deciding what must stay consistent together.
- - Use `packs/data-systems/core.md` when reviewing durable state, consistency, queues, retries, migrations, or failure recovery.
- - Use `packs/operations/core.md` when the work must survive production — timeouts, retries, load shedding, rollback, or alerting.
- - Use `packs/threat-modeling/core.md` when mapping trust boundaries, attacker paths, abuse cases, mitigations, or security verification.
- - Use `packs/ai-engineering/core.md` when building an LLM product, RAG system, model evaluation, or tool-enabled workflow.
- - Use `packs/agent-design/core.md` when building an agent or subagent, splitting work across agents, setting tool permissions, or deciding how agent output gets verified.
- - Use `packs/legacy-change/core.md` when modifying code that already works, has no tests, or is being replaced incrementally.
- - Use `packs/testing-design/core.md` when deciding what to test, what to mock, or why a suite is brittle or untrusted.
- - Use `packs/product-discovery/core.md` when the value of the thing itself is unproven — before the engineering packs, not alongside them.
- New packs follow the same overlay pattern. Consult `packs/registry.md` for pairings and for choosing between adjacent packs.
- ### 3b. Check whether a target grade is active
- A grade is different from a pack: it does not add topic depth, it sets the target the whole run is aiming for and changes what "done" means. Load one when the user names a target explicitly — "make this production ready", "MVP is fine", "get this to production grade", `$socratic mvp`, `$socratic production` — or when the request states it ("this needs to survive real traffic", "just a prototype for now").
- Consult `grades/registry.md`, load the matching grade file, and follow the domains and packs it marks mandatory in addition to whatever the dynamic scan already selected. When a grade is active, its gate — not step 5's generic sufficiency check — decides when the run stops. No grade named means no change: proceed as below.
- Grades are cumulative: `enterprise` supersedes `production` supersedes `mvp`. Resolve the entire chain down to `mvp`, not only the items listed in the named grade's own file — each file states what it supersedes and does not repeat those items in different words, so there is exactly one definition of each check to satisfy, not several that could disagree.
- ### 4. Self-answer the selected material questions
- Resolve each selected question in this order:
- 1. **Read first.** Use the codebase, configuration, documentation, prior conversation, and repository conventions.
- 2. **Apply the engineering default.** When evidence does not decide an engineering choice, take a defensible default and record consequential assumptions.
- 3. **Escalate only authority decisions.** Ask the user only about business priorities, budget, vendor choice, target market, legal risk tolerance, or irreversible decisions that materially change the result.
- Do this silently. The user should see the resulting decisions, not the raw question bank.
- ### 5. Run a sufficiency check and stop deliberately
- Socratic is not an instruction to exhaust a questionnaire. Its purpose is to reduce material uncertainty until the agent has a solid, evidence-backed basis to act.
- **If a grade is active, skip this step's five conditions and use the grade's gate instead** — stop when every gate item is resolved, mitigated with a stated reason, or marked not applicable, not when this section's conditions happen to be met. The rest of this step applies only when no grade was named.
- After each domain or decision cluster, stop expanding the review when all of the following are true:
- 1. The requested outcome, scope, and consequential assumptions are clear.
- 2. Every material risk has a mitigation, a verification step, an explicit acceptance by the appropriate authority, or a clearly stated escalation.
- 3. No unresolved contradiction changes the implementation plan.
- 4. The next plausible question would not materially change the design, risk, cost, authority decision, or verification plan.
- 5. The plan has a proportionate way to falsify its riskiest assumptions.
- If any condition fails, add the smallest relevant domain, deeper question depth, or pack. Do not continue merely to use more questions, and do not stop merely because a token budget is low. Stop because the remaining uncertainty is immaterial to the current task.
- ### 6. Emit the output contract once
- Before implementation, emit:
- ```text
- Domains considered: <each selected domain and why>
- Self-answered highlights: <5-10 decisions that shaped the design>
- Assumed (flag if wrong): <consequential defaults>
- Open questions for you: <ideally 0-3 authority decisions>
- Top risks: <material risks from the selected domains>
- Plan: <what will be built>
- ```
- Batch any genuinely blocking open questions. Do not turn the interrogation into the deliverable.
- ### 7. Build and verify
- Build after blocking decisions are resolved. Apply the Verification guidance from every loaded domain and pack, including anything added during the review. Report what passed, failed, or could not be verified without external access.
- ## Interactive mode
- When the user explicitly requests an interview:
- 1. Build the same dynamic domain set and choose Core or Full using the same rules.
- 2. Ask one material decision at a time with a recommended default.
- 3. On a correction, absorb it and continue. If the user says to stop or proceed, self-answer the remainder and build.
- 4. Keep the interaction proportional: 0-2 questions for a one-off, 3-6 for a prototype, 8-15 for production, and 15-25 for money, PII, health, or regulated systems.
- 5. Emit the same output contract before building.
- ## Preset domain combinations
- Use these as a sanity check, not a fixed router:
- | Building | Expected domains |
- |---|---|
- | CRUD web app | 00, 01, 02, 03, 05, 07 |
- | Public API or SDK | 00, 02, 04, 05, 08, 12, 07 |
- | Connectors or integrations | 00, 04, 05, 07, plus each connector's dependencies |
- | AI agent or chatbot | 00, 09, 05, 12, 08, 11, 07 |
- | RAG pipeline | 00, 09, 03, 12, 13, 07 |
- | Data pipeline or ETL | 00, 03, 06, 08, 13, 07 |
- | Mobile app | 00, 10, 01, 05, 11, 07 |
- | Infrastructure change | 00, 06, 08, 12, 14, 07 |
- | Payments feature | 00, 02, 03, 05, 13, 07 |
- | One-off script | 00, 07 |
- ## Guardrails
- - Do not ask what project inspection can answer.
- - Do not expose the full question bank to the user by default.
- - Do not escalate engineering decisions merely to avoid responsibility.
- - Do not keep interviewing after the user asks to proceed.
- - Do not claim verification without evidence.
- - Do not present legal conclusions; identify matters requiring counsel.
- - Do not depend on vendor-specific planning modes, memory systems, tool names, or subagent APIs.
README.md · README · 282 lines
View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)
- <p align="center">
- <img src="assets/pix.png" alt="Socratic Logo" width="380">
- </p>
- <h1 align="center">Socratic</h1>
- <p align="center">
- <img src="https://img.shields.io/github/license/m4vic/socratic" alt="License">
- <a href="https://github.com/m4vic/socratic/stargazers"><img src="https://img.shields.io/github/stars/m4vic/socratic" alt="Stars"></a>
- <a href="https://github.com/m4vic/socratic/network/members"><img src="https://img.shields.io/github/forks/m4vic/socratic" alt="Forks"></a>
- <img src="https://img.shields.io/github/last-commit/m4vic/socratic" alt="Last Commit">
- <img src="https://img.shields.io/github/v/tag/m4vic/socratic" alt="Version">
- </p>
- <p align="center">
- Meta - questioning about questioning. Question yourself till you're left with only answers.
- </p>
- **Socratic** is a self-interrogation skill for agentic AI, packaged as a portable Claude and Codex skill or prompt.
- It helps an LLM slow down, inspect the task, ask itself the right engineering questions, and only ask the user for the few decisions that actually require human authority.
- Two halves work together:
- - **697 questions** across 15 engineering domains — *what* a task must resolve
- - **60 decision cards** in 10 source-backed packs — *how* experienced engineers resolve it, distilled from Kleppmann, Nygard, Evans, Ousterhout, Feathers, Khorikov, and others
- Neither is loaded whole. The agent selects the slice the task actually signals, so a typical run costs about **3,000 tokens**.
- Not a form. Not a checklist. A lightweight reasoning loop.
- ## The stopping rule: question until the answer is solid
- Socratic is not a 697-question checklist and it is not an endless reasoning loop. It asks the smallest set of relevant questions needed to reach a solid, evidence-backed answer.
- The agent stops expanding its review when the outcome and scope are clear, consequential assumptions are visible, material risks have a mitigation or verification path, no contradiction changes the plan, and the riskiest assumption can be tested proportionately. If the next question would not change the design, risk, cost, authority decision, or verification plan, it is no longer material.
- This is the purpose of self-interrogation: let an agent challenge its own plan from several useful directions, satisfy the meaningful uncertainty, and then build. It should never keep thinking merely to consume tokens, nor stop only because a token budget is low.
- ## How it actually works
- **The agent interviews itself, not you.** Point it at a task, and it silently works through the relevant slice of the question bank — reading the codebase where it can, applying sensible engineering defaults where it can't, and only stopping to ask you about the handful of decisions that are genuinely yours to make (budget, vendor, legal risk, an irreversible call).
- What you see is a short contract before any code is written:
- ```text
- Domains considered: requirements, data, security, testing, AI/LLM
- Self-answered: Postgres over SQLite (existing driver, concurrent writers)
- at-least-once delivery; duplicate work is safe here
- no PII in the queue, so no encryption-at-rest work
- Assumed (flag if wrong): single-region; retries capped at 5
- Open questions for you: 1. Is a 30-second worst-case job latency acceptable?
- Top risks: retry storm on downstream outage — no circuit breaker yet
- Plan: durable enqueue → worker drain → backoff on failure
- ```
- One question instead of twelve. The other eleven were answered from your repo.
- The loop behind it:
- 1. classify the task
- 2. detect the relevant domains
- 3. load the smallest useful question set
- 4. self-interrogate silently
- 5. surface assumptions, risks, and only necessary open questions
- 6. build
- 7. verify
- If you'd rather be interviewed live, ask for it ("interview me," "ask me one at a time") and it switches to a one-yes/no-question-per-turn mode instead. That's opt-in, not the default.
- ## Core and Full modes
- The full question bank contains 697 questions across engineering domains.
- That is useful for production-grade, high-risk, or audit-style work, but too expensive to load by default for many agentic workflows.
- So Socratic now has two depths:
- ### Core
- Core is the default path.
- It loads the highest-signal subset of questions first: roughly the top 90 questions that matter most across routine, prototype, internal, and moderately scoped work.
- Use Core for:
- - prototypes
- - internal tools
- - one-off builds
- - medium-scope engineering work
- - most day-to-day agentic tasks
- ### Full
- Full loads the complete domain files.
- Use Full for:
- - production systems
- - public APIs
- - authentication
- - payments
- - PII or regulated data
- - autonomous tools
- - costly or irreversible actions
- - deep audits
- - explicit high-risk reviews
- ### Why this split matters
- The point is not to ask more questions.
- The point is to ask the most useful questions at the right time without burning unnecessary context.
- ### What it actually costs
- The whole repository is roughly 29,000 tokens. A normal run loads under a tenth of it. Counted with `tiktoken` (`o200k_base`), not estimated:
- | What loads | Tokens | When |
- |---|---:|---|
- | `SKILL.md` | ~2,300 | Every run |
- | **Typical pass** — SKILL + 4 core domains | **~2,800** | Most work |
- | All 15 core domains | ~4,100 | Broad prototype review |
- | Heavy pass — SKILL + 8 full domains + 1 pack | ~10,900 | Production, money, PII |
- | Entire repository | ~29,000 | Never |
- For scale: the median `SKILL.md` across Anthropic's 31 officially published skills is **2,255 tokens**. Socratic's entry point is the same size as a first-party skill — the depth lives in files that stay on disk until the task calls for them.
- **The Core/Full split is worth 7.5×.** All fifteen compact domain files together are 1,804 tokens; the fifteen complete ones are 13,599. Core files average about 120 tokens each, so even a maximal compact pass costs less than the skill's own instructions.
- ## Grades: run it once, land at a stated bar
- Packs add depth on one topic. A **grade** sets the target for the whole run — say "make this production ready" or `$socratic production`, and it changes what "done" means and when the run stops.
- | Grade | Bar |
- |---|---|
- | [`grades/mvp.md`](grades/mvp.md) | Does the one core action actually work, end to end, and fail honestly when it can't |
- | [`grades/production.md`](grades/production.md) | Would an on-call engineer accept ownership of this at 3am |
- | [`grades/enterprise.md`](grades/enterprise.md) | Would this pass procurement, an audit, and a real regional outage |
- Grades are cumulative — `enterprise` supersedes `production` supersedes `mvp` — and each file only states what it *adds*, never repeating the tier below it. That's deliberate: an item is defined in exactly one place, so loading several grade files together can't produce two different answers to the same question. `production` pulls in the `operations`, `threat-modeling`, and `testing-design` packs as mandatory and links to their cards rather than re-explaining them, adding only what nothing else covers — a tested restore, not just a configured backup; a stated latency target; a runbook; a kill switch. `enterprise` adds the organizational layer on top — a drilled disaster recovery, an audit trail, tenant isolation, a compliance sign-off path — the things a procurement review checks that engineering alone doesn't.
- No grade named, nothing changes — the normal sufficiency check runs as before and nothing in `grades/` loads.
- ## Knowledge-backed packs
- `questions/` and `packs/` have different jobs:
- - **Questions** provide coverage: they identify the concerns a task must resolve across requirements, security, data, testing, and other domains.
- - **Packs** provide depth: they add compact, source-backed reasoning for a specialist decision area after the relevant questions have been selected.
- Neither replaces the other. A great book does not cover every product, security, testing, or operational concern; a broad question bank does not contain every hard-won systems-design tradeoff.
- Ten packs ship today, each a handful of decision cards rather than a summary. [`packs/registry.md`](packs/registry.md) routes between them:
- | Pack | Depth it adds | Source |
- |---|---|---|
- | [`software-design`](packs/software-design/core.md) | complexity, interfaces, accidental generality | *A Philosophy of Software Design* |
- | [`domain-modeling`](packs/domain-modeling/core.md) | boundaries, aggregates, ubiquitous language | *Domain-Driven Design* |
- | [`data-systems`](packs/data-systems/core.md) | distributed data, consistency, migrations | *Designing Data-Intensive Applications* |
- | [`operations`](packs/operations/core.md) | timeouts, retries, load shedding, rollback, alerting | *Release It!*, Google *SRE* |
- | [`threat-modeling`](packs/threat-modeling/core.md) | trust boundaries, abuse paths, mitigations | *Threat Modeling*, *Security Engineering* |
- | [`ai-engineering`](packs/ai-engineering/core.md) | LLM evals, retrieval, tools, serving cost | *AI Engineering* |
- | [`agent-design`](packs/agent-design/core.md) | agent boundaries, tool permissions, verification | 34 shipped agents (empirical) |
- | [`legacy-change`](packs/legacy-change/core.md) | seams, characterization tests, incremental replacement | *Working Effectively with Legacy Code*, *Refactoring* |
- | [`testing-design`](packs/testing-design/core.md) | what to test, what to mock, why a suite is untrusted | *Unit Testing: Principles, Practices, and Patterns* |
- | [`product-discovery`](packs/product-discovery/core.md) | whether the thing should exist at all | *The Mom Test*, *Inspired* |
- For each task, the agent first selects the base domains and Core/Full depth, then reads the compact pack registry and adds zero to two relevant packs only where they sharpen the decision. Pack names describe the capability they add; their book sources are documented as provenance.
- Several packs overlap on purpose, so the registry says which to reach for: `software-design` covers module depth where `domain-modeling` covers where the boundaries fall; `data-systems` covers correctness of state under failure where `operations` covers staying available while failure happens; the Testing domain establishes what must be covered where `testing-design` decides whether the tests are worth keeping.
- `agent-design` is the one pack that isn't book-derived. Its cards come from structure observed across 34 agents shipped in first-party Claude Code plugins, where seven archetypes recur and the agent that does the work is never the one that verifies it. Structure observed in production beats structure argued from first principles.
- ### Books, compressed for agents — not summarized
- Socratic does not copy books into an agent context. It curates their reusable decision patterns into short cards: what to ask, the default answer, the tradeoff, the common mistake, when to escalate, and how to verify the choice. A pack is judged by whether it improves a real implementation decision — never by how much of the book it covers.
- ## What's in it
- | Domain | Questions | File |
- |---|---|---|
- | Requirements & scope | 40 | [`questions/00-requirements.md`](questions/00-requirements.md) |
- | Frontend & UI | 46 | [`questions/01-frontend.md`](questions/01-frontend.md) |
- | Backend & services | 45 | [`questions/02-backend.md`](questions/02-backend.md) |
- | Data & storage | 53 | [`questions/03-data.md`](questions/03-data.md) |
- | API design | 48 | [`questions/04-api.md`](questions/04-api.md) |
- | Security | 59 | [`questions/05-security.md`](questions/05-security.md) |
- | Infrastructure & DevOps | 46 | [`questions/06-infra.md`](questions/06-infra.md) |
- | Testing & quality | 38 | [`questions/07-testing.md`](questions/07-testing.md) |
- | Observability & ops | 39 | [`questions/08-observability.md`](questions/08-observability.md) |
- | AI / LLM / agents | 70 | [`questions/09-ai-llm.md`](questions/09-ai-llm.md) |
- | Mobile & offline | 41 | [`questions/10-mobile.md`](questions/10-mobile.md) |
- | Product & UX | 45 | [`questions/11-product-ux.md`](questions/11-product-ux.md) |
- | Cost & performance | 42 | [`questions/12-cost-performance.md`](questions/12-cost-performance.md) |
- | Compliance & legal | 42 | [`questions/13-compliance.md`](questions/13-compliance.md) |
- | Team & maintenance | 43 | [`questions/14-team-maintenance.md`](questions/14-team-maintenance.md) |
- Every file follows the same shape: **Priority 1** questions first, then thematic sections, then a **Verification** block to run *after* the build.
- ## Dynamic domain selection
- The domain set isn't chosen once from the initial request — it's built by scanning for signals and can grow mid-build. Requirements and Testing are always in. Everything else gets pulled in when it matches:
- | Signal | Domains added |
- |---|---|
- | UI, dashboard, form | Frontend |
- | service, job, queue | Backend |
- | database, schema, cache | Data |
- | API, SDK, webhook, **connector**, integration | API |
- | auth, payments, secrets, public-facing | Security |
- | deploy, CI/CD, cloud, scaling | Infra |
- | production, cron, monitoring | Observability |
- | AI, LLM, agent, prompt, RAG | AI/LLM |
- | mobile, iOS, Android, offline | Mobile |
- | anything user-facing | Product/UX |
- | scale, latency, high traffic | Cost/Performance |
- | personal data, health, EU/CA users | Compliance |
- | long-lived, team project | Team/Maintenance |
- A "tool with connectors" isn't just an API question — it pulls in API (contract, third-party auth), Security (credential storage per connector, blast radius if one leaks), and Testing (mocking each connector's failure modes) together. If self-answering later reveals a new need — say, a persistent store you didn't expect — the scan re-runs and adds Data mid-task.
- ## Three ways to use it
- ### 1. As a Claude Code skill
- ```bash
- mkdir -p ~/.claude/skills
- cp -r socratic ~/.claude/skills/
- ```
- ### 2. As a Codex skill
- Install under `$CODEX_HOME/skills`; when `CODEX_HOME` is unset, use `~/.codex/skills`:
- ```bash
- mkdir -p ~/.codex/skills
- cp -r socratic ~/.codex/skills/
- ```
- Invoke explicitly with `$socratic`, or let it trigger implicitly when your request matches.
- ### 3. As a portable prompt
- Paste [`PROMPT.md`](PROMPT.md) into the system prompt of any LLM — ChatGPT, Gemini, a local model, your own agent framework. Self-contained, no file dependencies.
- For an always-on prompt with minimal overhead, use [`PROMPT_LITE.md`](PROMPT_LITE.md).
- ## The self-interrogation loop
- ```
- Scan request + codebase → build working domain set (dynamic, can grow mid-build)
- ↓
- Self-answer every question: read codebase → apply engineering default → escalate only if it's a business decision
- ↓
- Emit contract once: domains / highlights / assumed / open questions (0-3 ideally) / risks / plan
- ↓
- Ask the (few) open questions, if any — batched
- ↓
- Build
- ↓
- Run Verification for every domain in the final set, including ones added mid-build
- ```
- ## Design principles
- - **Self-answer by default.** The bank exists so the agent has more engineering perspective, not so the user fills out a form.
- - **Read before assuming; assume before asking.** Escalation to the user is the last resort, reserved for decisions only they can authorize.
- - **Domain set is dynamic.** It's built from signals in the request and code, and can grow as the agent learns more mid-task — not fixed at the first guess.
- - **Testing runs every time.** Not gated behind the user asking for it — any tool/service/script that gets built gets its testing questions and verification pass.
- - **Keep "Open questions" near zero.** A long list of open questions means engineering decisions got escalated that shouldn't have been.
- - **Interactive mode is opt-in**, for when a user explicitly wants to be walked through it live.
- ## Extending it
- Add a domain by dropping `questions/15-yourdomain.md` in, following the existing shape (Priority 1 → sections → Verification), then add a row to the signal table in `SKILL.md` and `PROMPT.md` so it gets picked up dynamically.
- For source-backed specialist knowledge, add a pack under `packs/<name>/core.md` and optionally `packs/<name>/full.md`. Keep packs compact and decision-shaped rather than writing long summaries.
- ## License
- MIT — see [LICENSE](LICENSE). Use it, fork it, ship it.
- ## Contributing
- PRs welcome, especially:
- - Domains not covered (embedded, games, blockchain, hardware, accessibility-in-depth)
- - Questions that came from a real incident — those are the good ones
- - Better signal words for dynamic domain detection
Ratings and comments
Sign in to rate this skill.
Sign inFree to use. If it’s helped, pass it on.