More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Report Writing: Long-Form Research Reports

    1

    Build long-form reports through a gated pipeline — scope lock, separate research, claim-first outline with word budgets, cite-as-you-go drafting, and four ordered review gates.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Report Writing: Long-Form Research Reports

    Report Writing: Long-Form Research Reports

    Example session with this skill installed

    We need a report on whether our regional logistics firm should electrify its last-mile delivery fleet over the next three years. Management wants something they can actually decide from — maybe 6,000 words.

    • Read your context and instructions
    • Compiled the report writing

    STAGES 0–2 OUTPUT — produced by applying the skill (the pipeline serves the decision, so the demo shows the contract the rest of the report is built against).

    Scope note (signed off before research):
    Question: Should the firm electrify its last-mile fleet over 2027–2029, and if so, in what sequence?
    Reader: the operations director; decision: next year's vehicle procurement budget.
    Shape: ~6,000 words, 5 chapters, PDF deliverable.
    Out of scope: long-haul fleet; hydrogen alternatives; driver labor impacts.

    Research dimensions: (1) route and duty-cycle data — which routes fit current EV range in winter; (2) total cost per mile, EV vs diesel, including depot charging install; (3) vehicle and charger supplier landscape and lead times; (4) incentives and low-emission-zone regulation by city served. Cross-verification rule noted in each dimension's note: cost figures need two sources or get flagged single-sourced.

    Insights note (extracted, not left in the fact piles): the decision turns on duty cycle, not vehicle price — the firm's urban routes average 62 miles/day, well inside range, while 11% of routes exceed it; those routes, not the fleet average, drive the answer.

    Outline — briefs state claims, budgets sum to 6,000:

    1. Where the fleet actually drives (900) — claim: route data, not assumptions, decides which vehicles are candidates.
    2. Cost per mile, honestly loaded (1,400) — claim: EVs win on the urban duty cycle by year two even before incentives.
    3. Suppliers and lead times (1,000) — claim: charging installation, not vehicles, is the schedule risk.
    4. The 11% problem (1,200) — claim: the long routes need a hybrid answer, and pretending otherwise breaks the plan.
    5. Recommendation and sequence (1,500) — claim: electrify urban routes in 2027, hold the long-route decision to 2028 data.

    Drafting hasn't started — per the pipeline, it starts when this outline is approved.

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

    About this skill

    Long reports fail in predictable ways: research and writing interleaved until neither is done, chapters that contradict each other, citations bolted on at the end, an executive summary describing the report someone meant to write. This skill replaces that drift with a pipeline of six gated stages — scope, research, outline, draft, review, assemble — where each stage produces a file the next stage consumes, and a stage that hasn't produced its file isn't finished. It opens with an inputs/outputs contract and a scope lock: one-sentence question, named reader and the decision the report informs, length and shape, and an explicit out-of-scope list, signed off before any research starts. Research runs as its own stage, decomposed into dimensions, with load-bearing facts cross-verified or flagged single-sourced, and insights extracted into a separate note so the argument isn't buried in a pile of facts. The outline is a contract too: every chapter brief states its claim in one sentence (argument first, topics second) and carries a word budget, with budgets summing to the target before drafting begins. Drafting rules are the ones that actually save reports: one chapter per task, cite as you go with no exceptions, one shared style note. A chapter readiness rubric (claim delivered, budget, citations, voice fit — scored 0–2 each, under 6 of 8 goes back) makes quality checkable instead of vibes. Review runs four gates in order — section edit, coherence, bookends written last from the finished body, citation audit — and assembly validates hierarchy, cross-references, and numbering before any format conversion. It includes a scope note template, a chapter brief template, a stage-gate summary table, a worked scoping example, and a failure-mode table mapping each classic symptom to its cause and fix.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Writing & Content.

    • It structures rigor; it doesn't supply it. If the underlying research is thin, the pipeline will surface that as flagged gaps rather than fix it.
    • The gates add overhead that pays off at multi-chapter length; for a two-page memo the full pipeline is more ceremony than the document needs — the skill says so in its scope.
    • Voice consistency across many writers still depends on a real style note; the skill requires one but can't write a good one for you.

    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 buying a skill.

    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

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 24 days ago

    Permissions

    No special permissions declared or detected

    Listed24 days ago
    Updated21 days ago

    What's inside

    Frequently Asked Questions

    You can, and for short pieces that's fine. At multi-chapter length the single-pass approach produces the classic failures this skill targets: chapters that repeat setup and contradict each other, citations that don't survive checking, and a summary written from the prompt rather than the evidence. The pipeline exists because each of those failures is cheap to prevent at its stage and expensive to fix at delivery.

    Yes — enter at Stage 2. But the research still needs the Stage 1 treatment first: check that load-bearing facts are cross-verified or flagged, and extract the insights into their own note. Outlines built straight from fact piles produce annotated bibliographies, not arguments. The skill's stage-gate table shows exactly what each stage needs as input, so you can see what your existing material is missing.

    Strict enough to matter. Budgets must sum to the target before drafting, the chapter rubric scores budget adherence (within 10% earns full marks, over 25% off scores zero), and the failure table names "report keeps growing" as a budget problem, not a writing problem. A chapter that genuinely needs more room triggers a scope conversation — not silent expansion.

    New in Writing & Content