Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Consultant Research Pro — Evidence Behind Recommendations

    1

    Frame consulting decisions, organize evidence, test hypotheses and trace recommendations to defensible sources.

    $9.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Consultant Research Pro — Evidence Behind Recommendations

    Consultant Research Pro — Evidence Behind Recommendations

    Example session with this skill installed

    Illustrative decision: Client asks whether price is causing churn. Five interview notes mention price; two mention missing features. A retention table reports 12% churn for the higher-price plan and 8% for the lower-price plan, but sample sizes, time window, segments and collection method are unknown. Assess what we can conclude, give an evidence ledger and plan the next research steps. Do not invent causality or recommend a blanket price cut.

    • Read your context and instructions
    • Compiled the consultant research pro

    Decision brief

    Current evidence supports investigating price as a churn hypothesis, not concluding that price causes churn. Do not recommend a blanket cut from these inputs. Missing cohort, denominator and collection details materially limit interpretation.

    Evidence ledger

    Supplied evidenceWhat it supportsWhat it cannot establishVerification need
    Five interview notes mention pricePrice arose in the supplied notesFrequency in the whole customer population or causal effectFull notes, participant selection, unique respondents, wording and context
    Two notes mention missing featuresFeature gaps are another stated concernRelative importance or exclusive causeWhether notes overlap respondents and what features were meant
    Higher-price plan: 12% churnReported rate for an unspecified cohortComparable causal impact of priceDenominator, period, churn definition and segment mix
    Lower-price plan: 8% churnReported rate for an unspecified cohortThat a lower price would improve retentionSame definitions, observation windows and comparability

    Calculation and interpretation

    Reported gap: 12% − 8% = 4 percentage points. Relative to 8%, that is a 50% higher reported rate. These calculations do not resolve selection, measurement or confounding; uncertainty intervals cannot be calculated without sample sizes and an appropriate method.

    Competing explanations

    Price may contribute. Different buyer segments, usage, onboarding, feature expectations, contract periods or exposure windows may also explain the rates. These are hypotheses to test, not observed causes.

    Next research plan

    1. Define the decision, churn metric, cohort dates and target segments.
    2. Validate source tables, denominators, exclusions and plan/customer histories.
    3. Review original notes, coding and sampling; do not treat mentions as independent votes without checking.
    4. Compare like-for-like cohorts and examine product adoption and feature issues.
    5. If justified, design an ethically and commercially approved pricing test with success metrics and guardrails; do not assume results in advance.

    Recommendation

    Commission a focused churn diagnosis before changing prices. The executive statement should read: “Price concerns appear in supplied qualitative evidence; observed plan-rate differences need cohort validation and do not establish causality.” No interviews, browsing or experiments were conducted by this response. Sources remain the user-supplied scenario.

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

    About this skill

    Give recommendations a visible evidence chain

    Consultant Research Pro helps advisory teams, strategists and analysts turn business questions into structured research and decision support. It frames the decision, defines what must be known and prioritizes evidence sources.

    It organizes supplied research, documents and datasets; evaluates provenance and freshness; separates fact, inference and assumption; reconciles conflicts; challenges alternative explanations; and produces executive conclusions with uncertainty and next research steps. Browsing can support current public-source work only when available.

    It does not invent interviews, survey responses, proprietary access, market sizes, benchmarks, citations or competitor capabilities. Quantitative analysis requires appropriate data; weak evidence leads to qualified conclusions. Without tools it can build a research plan and evidence ledger from supplied material. Recommendations remain subject to domain-appropriate human review.

    Evidence may be incomplete, stale, biased or noncomparable. Cannot infer population behavior, statistical certainty or causality from insufficient inputs. No inaccessible proprietary research or fabricated sources. Current-source verification and quantitative processing depend on actual capabilities. Research supports decisions but does not guarantee outcomes or replace qualified high-stakes review.

    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

    Money-back guarantee

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

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    Permissions

    Read Files
    Write Files

    File Scopes

    Authorized
    task
    inputs
    and
    agreed
    output
    directory
    only

    Read only authorized research inputs, transcripts and datasets. Write draft evidence ledgers and reports in agreed output locations. Optional browsing, code, database and research-platform access requires real host tools and least-privilege permissions. No paid database access is bundled or assumed. Minimize confidential client data. Interviews, surveys, messages, database changes and external sharing require explicit authority and verified tools.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    No. It can organize supplied sources and draft a research plan. Without browsing it labels current external facts unverified rather than inventing research or citations.

    It evaluates source authority, provenance, definitions and freshness, separates interpretations, and records uncertainty and the next verification step instead of silently choosing a convenient answer.

    Only evidence can support causal claims. It tests alternatives and qualifies conclusions; correlations, anecdotes and weak samples do not become causal proof.

    New in Research & Analysis