Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Lanewright Intake: raw project input to cited facts

    1

    Turn documents, designs, chats, a folder or a single line into one cited evidence file with gaps and assumptions marked. Step one before an SRS and a sprint plan.

    Free

    Security scanned
    Lanewright Intake: raw project input to cited facts

    Lanewright Intake: raw project input to cited facts

    Example session with this skill installed

    Run Lanewright Intake on this one-line brief: "Integrated digital public services portal (Superapps)" (the client wrote it in Indonesian). The repository already has .claude/kit.json.

    • Read your context and instructions
    • Compiled the lanewright intake

    Real output of a run from that single line (2026-11-07), after a two-round interview. The brief was in Indonesian, so the skill wrote the document in Indonesian; translated into English for this page. First 45 of 148 lines of docs/requirements/intake.md.

    Intake: Superapps

    ItemValue
    ProjectSuperapps: an integrated digital public services portal
    Project kindNew application. The brief asks for a portal to be built and names no change request, bug list or running system; the repository holds no application code yet [S1; S2 "modules"]
    Document languageIndonesian
    Prepared2026-11-07
    Sources2
    Facts13
    Gaps16
    Assumptions16

    Sources

    #SourceTypeDateAuthorNotes
    S1"Integrated digital public services portal (Superapps)" (original in Indonesian)one-line briefunknown (received 2026-11-07)clientThe client's only input. Read in full; too thin on its own, so the skill moved on to the interview. Cited as [S1] because it has no pages or sections
    S2.claude/kit.jsonconfiguration (JSON)unknownunknownRead in full; cited per JSON key ([S2 "stack"]). Holds the stack, modules, branches, work regions, CI gates, document locations and forge

    Note: the user's request (prompt.md) is cited as [request]. out.json and err.txt are empty files from the test run, not sources.

    Facts

    Goals

    • [goal] The client asks for an "integrated digital public services portal" named "Superapps" [S1]
    • [goal] Project name in the repository configuration: superapps [S2 "project"]

    Actors

    • Gap G2: the users and their access rights are not named, see Gaps

    Features

    • [feature] The product is a digital portal for public services that are "integrated" [S1]

    Business rules

    • Gap G14: the rules per service (requirements, processing time, status flow) are not named, see Gaps

    Data

    • Gap G9: the data owned and shared is not named, see Gaps

    …

    Next step: Lanewright SRS turns this intake into a requirements document.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Lanewright Intake reads whatever raw material a project starts with and writes one file, docs/requirements/intake.md, where every fact has a kind (goal, actor, feature, business rule, data, screen, integration, constraint, priority, deadline, budget) and a citation to where it came from.

    It accepts PDF, DOCX, Markdown, plain text, images and screenshots, chat exports, a local folder, a Google Drive folder (through your agent's Drive connector), pasted text, or a single line such as a project title. DOCX files are read by a small bundled Python script; PDFs and images are read by the agent, with fallbacks when it cannot.

    What you get:

    • A sources table (file, type, date, author, how it was read)
    • Facts grouped by kind, each with a citation such as [S2 p.7] or [interview Q3]
    • A gap list: what an SRS and a sprint plan need that the sources do not say
    • At most two short interview rounds with a proposed default for every question, including "which 3-5 features must exist at launch?" Anything still unknown is written as an ASSUMPTION, never as a fact
    • The project kind, new application or change to an existing system, which the SRS skill uses

    The document follows the language of your sources. It never invents a fact, never copies hosts, credentials or personal data into the output, and never overwrites a file without asking.

    Next step: Lanewright SRS turns the intake into a requirements document. Free, no accounts, Python 3.8+.

    These four skills prepare the work. To run the sprints, use Lanewright (paid): https://www.agensi.io/r/HUZAIRSAPUTRA?to=/skills/lanewright-sprint-workflow-for-ai-coding-agent-teams or the free scrum-kit Claude Code plugin: https://github.com/huzairsaputra/scrum-kit. They work without either.

    Why four skills, not one: https://github.com/huzairsaputra/lanewright-planning#why-four-skills-not-one

    The four free skills on Agensi: Intake https://www.agensi.io/r/HUZAIRSAPUTRA?to=/skills/lanewright-intake-raw-project-input-to-cited-facts, SRS https://www.agensi.io/r/HUZAIRSAPUTRA?to=/skills/lanewright-srs-cited-requirements-in-word-and-markdown, Sprint Planner https://www.agensi.io/r/HUZAIRSAPUTRA?to=/skills/lanewright-sprint-planner-srs-to-weekly-plan-and-backlogs, Backlog Check https://www.agensi.io/r/HUZAIRSAPUTRA?to=/skills/lanewright-backlog-check-readiness-check-and-human-sign-off

    Quality depends on the source; scanned images need an agent with vision; audio and video are not supported yet (transcripts work; direct audio is planned through a speech-to-text service).

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 4 days ago

    • Free with an Agensi account

    Requires

    Python 3.8+

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    File Scopes

    docs/requirements/**
    Listed4 days ago

    What's inside

    Frequently Asked Questions

    It interviews you for the 3-5 must-have features, team and deadline, with a default for each, and writes everything it could not confirm as an ASSUMPTION.

    Only if your agent can read images; otherwise they are listed as a gap, never guessed.

    The skill sends nothing anywhere; your agent does whatever your agent does.

    New in Business & Operations