Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Execution Evidence Handoff Pack

    1

    Convert delegated task outputs into verifiable handoff packs with explicit evidence ledgers and recovery paths.

    $9.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Execution Evidence Handoff Pack

    Execution Evidence Handoff Pack

    Example session with this skill installed

    Review the handoff for the database migration. I have the migration log showing 200 OK, a screenshot of the new schema in the UI, and a Jira approval from last month for the migration plan.

    • Read your context and instructions
    • Compiled the execution evidence handoff

    The execution handoff for the database migration is

    CONDITIONALLY_READY. While the migration action is evidenced, the approval context is stale and the UI screenshot provides only indirect proof of system state.

    A. Executive decision

    • Verdict: CONDITIONALLY_READY
    • Reason: The Jira approval links to an older version of the migration plan (Stale).
    • Risk: Potential mismatch between approved schema and executed schema.
    • Safe to accept: The structural changes are confirmed via logs, but the authorization needs re-validation.

    F. Execution Evidence Transfer & Acceptance Ledger (EETAL)

    Action IDClaimed OutcomeEvidence StateFreshnessIdentity MatchHandoff Decision
    ACT-001Schema MigrationCONFIRMEDCURRENTEXACT_MATCHACCEPT_EVIDENCE
    ACT-002Data IntegrityPROBABLECURRENTUNKNOWNREPERFORM_VALIDATION
    ACT-003AuthorizationSTALESTALEMISMATCHREPAIR_EVIDENCE

    J. Recovery path

    1. Action: Re-link the current migration script hash to a new Jira comment or approval occurrence.

    Action: Run a checksum validation on the production table users_v2 to confirm row count matches the pre-migration snapshot.
    3.

    Action: Retrieve the direct system receipt from the migration tool's internal audit table instead of relying on the UI screenshot.

    L. Reviewer verification procedure

    1. Locate object: Access the production DB instance and query information_schema.tables.
    2. Confirm identity: Verify table users_v2 exists and matches the version defined in migration_v1.0.4.sql.
    3. Inspect state: Confirm the last_login_at column is present and nullable.
    4. Verify approval: Check Jira ticket PROJ-123 for an approval timestamp within 24 hours of the execution log.

    Next steps

    • Update the EETAL with the current approval ID from Jira.
    • Run the provided SQL checksum script to move ACT-002 to CONFIRMED.
    • Sign off on the handoff pack once the stale authorization gap is closed.

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

    About this skill

    Delegated tasks often end with weak handoffs like "done" messages, vague screenshots, or logs that don't confirm actual system state. This skill transforms loose workflow evidence into a rigorous, reviewer-verifiable handoff pack that separates confirmed proof from narrative claims.

    What it does

    • Reconstructs execution claims by mapping material actions to specific, verifiable system evidence.
    • Assesses evidence quality using strict states like CONFIRMED, STALE, or CONTRADICTED to prevent false success reporting.
    • Identifies object identity to ensure that approvals and results apply to the correct version or occurrence.
    • Generates a recovery path providing specific repair actions for missing or insufficient proof instead of blind retries.
    • Produces an EETAL ledger containing 26 data points per claim, including source locators and reviewer reproduction steps.

    How it works

    1. Inventory phase: You provide the task objective, tool logs, receipts, screenshots, and any output artifacts.
    2. Analysis phase: The skill links every claimed outcome to its specific evidence, checking for freshness and directness.
    3. Ledger generation: It builds the Execution Evidence Transfer & Acceptance Ledger (EETAL) and a delegation authority map.
    4. Readiness verdict: It delivers a final status (e.g., READY_FOR_ACCEPTANCE or CONDITIONALLY_READY) with clear reviewer instructions.

    Frameworks & tools

    This tool is platform-agnostic, working with any stack that generates logs, transaction IDs, or artifacts. It is particularly effective for systems using GitHub, Jira, AWS, Stripe, or custom REST APIs where side-effect verification is critical.

    Why this beats prompting it yourself

    Generic prompts often accept "success" messages at face value without checking if the side effect actually persisted. This skill applies a control-system logic that treats missing evidence as a blocker, preventing high-consequence failures from being hidden by narrative summaries.

    Use cases

    • Multi-agent handoffs where one agent must prove its work to a human or a supervisor agent.
    • Audit-ready releases where every code change requires linked test results and deployment receipts.
    • External vendor verification to validate that third-party API mutations or manual tasks were performed correctly.
    • Compliance reporting for workflows that demand strict chain-of-custody and version matching for artifacts.

    Known limitations

    This skill does not perform live system actions or create tool access. It cannot certify legal or financial compliance and relies entirely on the accuracy of the logs and artifacts you supply.

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

    No live system execution. Requires external logs/artifacts. Does not infer state from missing logs. Operational verification only, not legal/financial certification.

    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 1 month ago

    Permissions

    Terminal / Shell
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This skill standardizes the handoff process by transforming vague 'task complete' messages into a rigorous ledger of verified system states and evidence links, preventing silent failures in delegated work.

    It is platform-agnostic and compatible with any AI agent or workflow engine that can export execution logs, API response bodies, or artifact metadata, such as AutoGPT, LangChain, or custom-built agents.

    You receive the core skill integration for your agent environment, the EETAL (Execution Evidence Transfer & Acceptance Ledger) schema, and structured templates for generating recovery paths and reviewer instructions.

    No, this skill functions as a post-execution analysis layer; it processes the logs and artifacts you provide to verify outcomes without requiring direct access to your internal tools or production systems.

    If the skill detects that an outcome isn't backed by fresh or direct evidence, it generates a 'Recovery Path'—a specific set of instructions or repair actions needed to satisfy the verification requirements.

    Installation is straightforward: you simply import the skill into your agent's workflow and configure it as the final evaluation step before a task is marked as accepted or handed off to a human.

    New in Agents & Orchestration