Borrowing it
Nothing to install: this file belongs to radekamirko/C.R.I.S.P. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/radekamirko/C.R.I.S.P/main/.claude/skills/phase3-investigate/SKILL.mdgit clone --depth 1 https://github.com/radekamirko/C.R.I.S.PWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/radekamirko/c.r.i.s.p/phase3-investigate)<a href="https://agentmods.dev/skills/radekamirko/c.r.i.s.p/phase3-investigate"><img src="https://agentmods.dev/badge/skills/radekamirko/c.r.i.s.p/phase3-investigate/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/radekamirko/c.r.i.s.p/phase3-investigate"><img src="https://agentmods.dev/badge/skills/radekamirko/c.r.i.s.p/phase3-investigate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00093 | $0.05175 |
| Opus 5 | $0.00046 | $0.02587 |
| Sonnet 5 | $0.00019 | $0.01035 |
| Haiku 4.5 | $0.00009 | $0.00517 |
Grade A, and why
phase3-investigate 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 12d ago.
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.
How it starts
The opening of the file, as written. The whole thing — 432 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pacing Rule — One Section at a Time
Present one section, deliverable, or elicitation move at a time. After presenting — stop. Wait for the client to respond. Do not pre-fill and present multiple sections in one message. Do not move to the next step until the client confirms, corrects, or gives a clear go-ahead.
The goal is a conversation, not a document dump. If you've written more than one section without a client response in between — you've gone too far.
I — Investigate: Process Understanding
Map what exists (or what should exist) before designing a solution.
Project State
At the start of Phase I: read
docs/crisp-state.json. Check phases C and R are complete. Useproject.type,phases.R.hitlZones,phases.R.stakeholdersfor context.At the end of Phase I (before exit checklist): update
docs/crisp-state.json:
- Set
phases.currentto"S"- Add
"I"tophases.complete- Set
phases.I.completetotrue- Fill
phases.I.processTrack,phases.I.userTypes,phases.I.dataMappingRequired,phases.I.uxDiscoveryRequired,phases.I.navigationPattern- Add any unresolved items to
phases.I.openQuestions
Two Tracks
Track A: Existing Process (Automation)
Shadow the actual person doing the work. Step by step, no skipping.
- What tool is open at each step?
- What is copied, pasted, where does it go?
- What decisions are made manually and why?
- Where do errors happen? Where do things slow down?
- Every tab, every click, every "and then I just..."
Goal: expose the real process, not the documented one. They are always different.
War story: Client wanted drag-and-drop category sorting for her e-commerce shop. A week of dev work, minimum. I asked why. "I just want to change the order for the season." I asked how she wanted to reorder it. She told me. Dev made the change in 4 minutes — one config update. The drag-and-drop feature never got built. Shadow the real process before you spec anything. The ask is almost never the need.
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.
- 12d ago First seen · 432 lines · 0 tokens per session scan A affb996a2eb8
phase3-investigate is a skill published in the GitHub repository radekamirko/C.R.I.S.P (52 stars, last pushed 3mo ago), licensed MIT. It adds 93 tokens to every session and 5,175 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
accessibility
Audits code for WCAG 2.2 AA compliance and provides design guidance for accessible components. Covers semantic HTML, ARIA, keyboard nav, contrast, focus management, and motion.
abd-design
ABD Design agent: produces architecture diagrams, component breakdowns, data flow, and design decisions from the current plan.
design-tokens
This skill should be used when a user wants to scaffold or refactor a design token system — e.g. "set up design tokens", "create a design system foundation", "extract these hardcoded colors into tokens", "add a type scale", "theming setup", or "dark mode tokens". It produces a single source of truth for color, type…
accessibility-audit
This skill should be used when evaluating a design, page, or component against accessibility standards and producing a severity-scored report. Trigger phrases include "audit accessibility", "check a11y", "is this accessible", "WCAG audit", "check color contrast", and "accessibility review before handoff". Use it to…
accessible-components
This skill should be used when the user wants to build or fix accessible UI components or set up accessibility tooling — triggers like "make this modal accessible", "build an accessible dropdown", "add ARIA to this", "accessible tabs/combobox/tooltip/accordion", "set up a11y linting", or "fix keyboard navigation". It…
interaction-feedback
This skill should be used when the user wants clear visual feedback for async actions and state changes — triggers like "add loading and error states", "show a success message", "add a toast/notification", "inline form validation", "optimistic UI", "empty state for this list", "this button gives no feedback", or "add…