feedback verify with real data

    1

    Enforce codebase and API state verification using grep and data probes to eliminate assumptions and hallucinations.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    feedback verify with real data

    Example session with this skill installed

    Verify if the FMP API key is on the Starter or Premium tier and check if the BatchSyncEngine.available() method is implemented in the backend.

    • Read your context and instructions
    • Compiled the feedback verify

    I have performed a codebase search and a data-driven API probe to verify the current state.

    Verification Results

    TargetMethodResultEvidence
    API TierFMP Daily Bar (2015-01-02)StarterEmpty array returned for data > 5 years old.
    Symbol Existencegrep -rn "def available"Not FoundNo match in backend/app/.
    Backend Routergrep for @router.getVerifiedbackend/app/routers/sync.py:42

    The BatchSyncEngine.available() method does not currently exist in the codebase. You will need to implement it before it can be used by the rate limiter. The FMP API key is behaving as a Starter tier key because it failed to retrieve historical data from 2015, which is a known limitation of that tier.

    Next steps

    1. Implement the available() method in backend/app/engines/sync_engine.py.
    2. Upgrade the FMP API key to Premium if access to 5+ years of historical data is required.
    3. Run grep -rn "BatchSyncEngine" to identify all call sites that need updating.

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

    About this skill

    Stop relying on assumptions, outdated documentation, or hallucinated function signatures. This skill enforces a "verify first" workflow, requiring concrete evidence from the codebase or live API responses before making any claims. It prevents the common pitfall of assuming a feature exists just because it is mentioned in a README or a TODO list.

    What it does

    • Symbol verification ensures functions, classes, and variables actually exist in the current directory before referencing them.
    • API tier probing uses data-driven tests to determine actual service limits rather than relying on HTTP status codes.
    • TODO consolidation filters out completed tasks by cross-referencing checklist items against the live codebase.
    • Endpoint mapping validates the existence of specific routes using targeted grep patterns on router files.
    • Uncertainty handling provides exact commands for the user to run when local tool access is restricted.

    How it works

    1. Grep codebase for specific function definitions, symbol references, or migration keywords before asserting state.
    2. Execute data probes against external APIs by requesting specific data ranges or schemas that differentiate between service tiers.
    3. Cross-reference tasks against the existing file structure to identify and drop obsolete TODO items.
    4. Cite evidence by providing file paths and line numbers for every claim made about the system.

    Frameworks & tools

    This skill utilizes standard shell utilities like grep, ls, and bash. It is designed for developers working with REST APIs and multi-directory codebases across backend/, frontend/, and db/migrations/ structures.

    Why this beats prompting it yourself

    Standard prompts often lead to confident hallucinations based on outdated context or docstrings. This skill mandates a specific verification loop using grep and data-driven probes, ensuring every statement is citable to a specific line of code or a validated API response shape.

    Use cases

    • Auditing a legacy codebase to find orphaned symbols or missing endpoints.
    • Determining if an external API key has exceeded its quota or is restricted to a lower tier.
    • Cleaning up stale TODO lists by verifying which features have already been merged.
    • Validating the success of a database migration by checking the migration history.

    Known limitations

    Verification is limited to the files and tools accessible in the current environment. It cannot verify external API behavior without an active network connection or valid API keys.

    Requires grep and bash access. API probing requires live network connectivity and valid keys. Scope is limited to the provided directory structure.

    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

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 28 days ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Read Files
    Network Access
    Listed28 days ago

    What's inside

    Frequently Asked Questions

    This skill enforces a 'verify first' workflow by using grep and data probes to check for actual code symbols and API responses, preventing the agent from relying on outdated documentation or hallucinated function signatures.

    The skill is designed for any AI agent with access to a terminal or shell environment (like ChatGPT with Analysis, Claude with Code Interpreter, or local IDE agents) capable of running grep and bash commands.

    Yes, the skill includes 'API tier probing' logic that uses data-driven tests to verify service limits and schema shapes, ensuring the agent knows the true state of your external integrations.

    You will receive the specialized prompt templates and configuration logic required to enable the verification loop, along with instructions for integrating it into your agent's workflow.

    Unlike standard prompting, this skill mandates a systematic evidence-based loop that requires file paths and line numbers for every claim, ensuring the agent never assumes a feature exists just because it's in a README.

    New in Prompt & Skill Engineering