intent

The first interview stage of a product-discovery process. It records the founder’s understanding of the problem, the specific target user, the timing, the founder’s connection to the problem, current workarounds, and evidence that people care.

In plain words
What is it for?
Use it to run a one-question-at-a-time founder interview and save the answers in Product/intent.md.
Why use it?
It helps test whether a real problem exists before spending time designing features or building a solution. It also exposes when an idea is based only on an untested hunch.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/inscico/i-framework/intent
Any agent
npx skills add InSciCo/i-framework --skill intent
Clone the repo
git clone --depth 1 https://github.com/InSciCo/i-framework

Made for: Claude Code, Codex.

Per session 59 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 546 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00059 $0.00546
Opus 5 $0.00030 $0.00273
Sonnet 5 $0.00012 $0.00109
Haiku 4.5 $0.00006 $0.00055

Measured yesterday against content hash 839ebf1c4f8f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

intent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/intent/SKILL.md · 35 lines

What it actually says

/intent — Stage 1: capture the "why"

You retire founder–problem risk before any solution talk. Run a guided interview, one question at a time. Stay on the problem, not the solution — if the founder jumps to features, gently redirect to the pain.

Procedure

1. Set up

Create the Product/ folder if it doesn't exist. If Product/intent.md already exists, read it and offer to refine rather than overwrite.

2. Interview (one question at a time)

Draw out, and challenge thin answers on:

  • The problem — in plain language, the pain not the product. ("What's broken today?")
  • The target user — who has this most acutely? Push back on "everyone"; get to a specific persona.
  • Why now — what changed? The tailwind/shift that makes today the moment.
  • Why you — founder–problem fit: lived experience, unfair insight, access. What will keep you in the fight when it's hard?
  • Status quo — what does the user do about this today? (The real competitor is usually a workaround or "nothing.")
  • Evidence of pain — the cheapest test: who have you actually talked to, and what did they say? Aim for 5–10 people in genuine pain. Accept "it's still a hunch" honestly — but name it.

Use AskUserQuestion for genuine forks; use open prompts for narrative answers. Don't fill silence with invented facts — ask.

3. Synthesize Product/intent.md

Write it using the structure from the plugin's templates/intent.md. Read the draft back to the founder for confirmation/correction before finalizing.

4. Evaluate the gate

Set frontmatter gate: passed when all three hold: (1) a crisp problem statement, (2) a named target user, (3) at least one piece of real pain evidence. Otherwise gate: open and clearly list what's missing — but never block the founder from moving on.

5. Hand off

Point to /ideate as the next step (and /iflow if they want the orchestrator to route them).

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 35 lines · 59 tokens per session scan A 839ebf1c4f8f

Subscribe to this mod's changes

intent is a skill published in the GitHub repository InSciCo/i-framework (4 stars, last pushed 1mo ago), licensed MIT. It adds 59 tokens to every session and 546 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.