More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Talley Outcome Packet Compiler

    1

    Turn explicit goals, dependencies and acceptance checks into a deterministic JSON plan. Catch cycles and missing references, preserve unverified claims, and get a next-action queue without executing tasks.

    $29

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Talley Outcome Packet Compiler

    Talley Outcome Packet Compiler

    Example session with this skill installed

    {
      "schema": "dow-input.v1",
    

    "outcome": "Prepare a synthetic office supply count for review.",
    "completion_condition": "A reviewer can reproduce the fixture totals and sign off the draft count.",
    "facts": [{"id":"f-fixture","statement":"This example uses invented stock records.","source":"synthetic example author"}],
    "constraints": ["Use synthetic records only.","Do not change any live inventory."],

      "goals": [{"id":"g-count","description":"Produce a checkable count from a fixed fixture."}],
      "milestones": [
        {"id":"m-review","goal_id":"g-count","title":"Review the draft count","depends_on":["m-total"],"acceptance_checks":["Reviewer matches every draft line to the fixture."],"todos":[{"id":"t-review","action":"Compare the draft count to the fixture.","actor":"reviewer","prerequisite":"Draft count and frozen fixture","expected_evidence":"Recorded line-by-line comparison","completion_check":"All lines match or have an explicit discrepancy.","risk":"Stop if a fixture line is missing."}]},
        {"id":"m-total","goal_id":"g-count","title":"Compute fixture totals","depends_on":[],"acceptance_checks":["A second calculation reproduces the totals."],"todos":[{"id":"t-total","action":"Calculate the total for each fixture item.","actor":"analyst","prerequisite":"Frozen synthetic stock fixture","expected_evidence":"Draft item totals","completion_check":"Every fixture item has one total.","risk":"Stop if item identifiers are duplicated."}]}
      ],
      "challenge": {"assumptions":[{"statement":"Fixture item identifiers are unique.","test":"Count duplicate identifiers before totaling.","halt_condition":"Any duplicate identifier is found."}]},
    

    "verification_requirements": ["Recalculate totals independently before claiming the count is correct."]
    }

    • Read your context and instructions
    • Compiled the talley outcome packet
    {
      "challenge": {
        "assumptions": [
          {
    
        "halt_condition": "Any duplicate identifier is found.",
        "statement": "Fixture item identifiers are unique.",
        "test": "Count duplicate identifiers before totaling."
    
          }
        ]
      },
      "compatibility": {
        "native_v12_task_packet": false,
    
    "scope": "local planning schema only"
    

    },
    "completion_condition": "A reviewer can reproduce the fixture totals and sign off the draft count.",

      "constraints": [
        "Use synthetic records only.",
    
    "Do not change any live inventory."
    
      ],
      "execution_authority": "not_requested",
      "facts": [
        {
          "id": "f-fixture",
          "provenance": "user-reported",
          "source": "synthetic example author",
    
      "statement": "This example uses invented stock records.",
      "verification_status": "unverified"
    
        }
      ],
      "goals": [
        {
    
      "description": "Produce a checkable count from a fixed fixture.",
    
          "id": "g-count",
          "status": "not started"
        }
      ],
      "input_sha256": "4aba02348e6f7e6d3fa02a27f81283bc0221608fcb20992264ae48361123f284",
      "milestones": [
        {
          "acceptance_checks": [
    
        "A second calculation reproduces the totals."
    
          ],
          "depends_on": [],
          "goal_id": "g-count",
          "id": "m-total",
          "status": "not started",
          "title": "Compute fixture totals",
          "todo_ids": [
            "t-total"
          ]
        },
        {
          "acceptance_checks": [
    
        "Reviewer matches every draft line to the fixture."
    
          ],
          "depends_on": [
            "m-total"
          ],
          "goal_id": "g-count",
          "id": "m-review",
          "status": "not started",
    
      "title": "Review the draft count",
    
          "todo_ids": [
            "t-review"
          ]
        }
      ],
    

    "outcome": "Prepare a synthetic office supply count for review.",

      "packet_sha256": "c443d1bda138e8c622d3fe69c2c4bc9f2d6b1aeed96b0e0cd6ada9366187c6cc",
      "production_admission": false,
      "schema": "dow-planning-packet.v1",
      "status": "not started",
      "todo_queue": [
        {
    
      "action": "Calculate the total for each fixture item.",
    
          "actor": "analyst",
          "blocked_by": [],
    
      "completion_check": "Every fixture item has one total.",
    
          "expected_evidence": "Draft item totals",
          "goal_id": "g-count",
          "id": "t-total",
          "milestone_id": "m-total",
    
      "prerequisite": "Frozen synthetic stock fixture",
      "rank": "now",
      "risk": "Stop if item identifiers are duplicated.",
      "status": "not started"
    
        },
        {
    
      "action": "Compare the draft count to the fixture.",
    
          "actor": "reviewer",
          "blocked_by": [
            "m-total"
          ],
    
      "completion_check": "All lines match or have an explicit discrepancy.",
    
          "expected_evidence": "Recorded line-by-line comparison",
          "goal_id": "g-count",
          "id": "t-review",
          "milestone_id": "m-review",
    
      "prerequisite": "Draft count and frozen fixture",
      "rank": "blocked",
      "risk": "Stop if a fixture line is missing.",
      "status": "not started"
    
        }
      ],
      "verification_requirements": [
    
    "Recalculate totals independently before claiming the count is correct."
    
      ]
    }
    

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

    About this skill

    Make an explicit plan inspectable

    Compile goals, milestone dependencies, acceptance checks and verification requirements into one bounded JSON planning packet. The compiler orders milestones, catches dependency cycles and missing references, and returns a queue labeled now, next or blocked. Supplied facts remain user-reported and unverified; every work status stays not started.

    What you receive

    A reusable agent skill, an unchanged standalone Python compiler from the existing Talley workflow, the precise input/output contract and a synthetic example. No third-party Python dependencies, account signup, API key or network connection is needed to compile.

    Where it stops

    Provide explicit milestones and acceptance criteria. The skill does not invent missing requirements, execute the plan, grant permissions, verify external facts or claim project completion. It includes no production control-plane integration or live MCP service.

    Verification scope

    The reused compiler passed 28 existing compiler tests. Seven additional local delivery checks passed, including execution from an isolated temporary file, deterministic output, unchanged source, no compiler-created files and rejection of an injected authority field. These are local checks; marketplace approval and live agent integrations are unverified.

    Documentation and demonstration

    Read the complete usage and field contract at https://talley-outcome-packet-docs.matthew946131.chatgpt.site. The demo video and six walkthrough images show captured output from a local synthetic CLI run. They are edited presentations of real local results, not recordings of live agent or Agensi runtime execution.

    Support

    Contact Matthew Talley through Agensi creator messaging with the skill version, Python version, exit code, exact error, and sanitized minimal input. Do not send credentials or confidential data.

    Input is a strict schema; incomplete free-form requests need explicit fields before compilation.
    Planning ranks reflect dependencies, not externally verified readiness.
    Hashes bind input/output representation; they do not prove truth, authorship or approval.
    No downstream execution, customer outcomes, native V12 mapping or independent security audit is included.
    Walkthrough images and video use synthetic input and captured local CLI results; individual agent integrations and live Agensi runtime execution are unverified.

    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 5 days ago

    Requires

    Python 3.9+
    Permitted local terminal and file tools
    Python

    Permissions

    File Scopes

    User-designated
    JSON
    input
    Invocation-specific
    temporary
    compiler
    and
    files

    Read the user-designated JSON input. Agent creates and cleans temporary source/input files. Run the bundled local Python compiler. No network access or downstream execution.

    Listed5 days ago

    What's inside

    Frequently Asked Questions

    A reusable agent skill, standalone Python compiler, exact input/output contract, and synthetic example. The package contains five text files.

    No. It returns planning data. Work remains not started, production_admission is false, and execution_authority is not_requested.

    Python 3.9+ and permitted local terminal/file tools. The compiler uses the Python standard library only.

    No API key, network service, or paid Agensi tool is required by the compiler. Your agent or host may have separate costs.

    The input is a strict JSON schema. An agent can help collect the explicit required fields, but missing acceptance criteria and dependencies must be supplied rather than invented.

    No. Supplied facts remain user-reported and unverified. Input claiming approval grants no authority, and hashes do not prove truth or permission.

    The CLI exits with code 2, returns an invalid_input JSON error on stderr, and returns no packet on stdout. Correct only the intended input; do not weaken criteria to force a pass.

    Identical valid input produces identical CLI output and digests. The local demonstration includes two matching runs.

    The standalone temporary-file CLI route is locally tested. Individual agent integrations and live Agensi runtime execution remain unverified. Native V12 mapping and a live MCP service are not included.

    Message Matthew Talley through Agensi creator messaging. Include version 1.0.0, Python version, exit code, exact error, and sanitized minimal input. No guaranteed response time or custom integration service is included.

    New in Workflow Automation