agentleFS
muhamedroshanmuhamedroshan/Odoo-Agent-Skills-Functional/skills/odoo-requirements-analysis/SKILL.md

odoo-requirements-analysis

Skill2 starsChanged 14 days agoMITView on GitHub

Turn raw client requirements into numbered, atomic Odoo requirements, generate clarification questions per business area (accounting, sales, purchase, inventory, manufacturing, HR, POS, website, projects, access rights, integrations, data migration) and confirm Must Have vs Nice to Have priorities with the user. Never assumes; every unknown becomes a question.

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Only ask questions that are relevant and not already answered.

    line 23
  2. Very specific:

    STOP and wait: Do not continue with a guessed default.

    line 30
  3. Distinctive:

    Ask the user to confirm each requirement as:

    line 33

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Odoo Requirements Analysis

Step 1: Atomize

For every statement the client makes, create one row in docs/01_requirements.md:

IDAreaClient wording (verbatim)Interpreted requirementSourcePriorityStatus
REQ-001Sales"we give discounts by customer"Customer-specific pricingEmail 2026-10-14TBCOpen
  • One requirement = one testable behaviour. Split compound sentences.
  • The Interpreted requirement column must not add facts. Anything interpreted becomes a question.

Step 2: Generate questions

For each area mentioned (and for the mandatory areas: Company/Localization, Accounting and Users/Access, which are always reviewed), read the matching bank in references/questions/:

  • company_localization.md, accounting.md, sales_crm.md, purchase.md, inventory.md, manufacturing.md, hr_payroll.md, pos_ecommerce.md, project_services.md, users_access.md, integrations_migration.md

Only ask questions that are relevant and not already answered. Log each one in 02_open_questions_and_decisions.md:

| Q-ID | Area | Question | Why it matters | Options / Odoo default | Blocking | Answer | DEC-ID |

Step 3: Ask (never assume)

  • Present the questions to the user grouped by area, at most 10 per batch, blocking questions first.
  • For each question, say what the Odoo standard behaviour is, so the user can simply answer "standard is fine".
  • STOP and wait. Do not continue with a guessed default. If the user says "you decide", write your recommendation as a decision marked Proposed, and get an explicit OK.

Step 4: Prioritize

Ask the user to confirm each requirement as:

  • Must Have: go-live is impossible or not accepted without it (legal, core operations, client's stated deal-breakers).
  • Nice to Have: improves efficiency or reporting; there is a manual or standard workaround.

You may propose a priority with a reason, but mark it Proposed until the user confirms.

Output checklist

  • Every client statement maps to at least one REQ.
  • Every REQ has a confirmed priority.
  • No blocking question is open.
  • Decisions are dated and attributed (who answered).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.