Independent methods. Inspectable results. Agent skills, local tools and weekly intelligence.

A reusable toolkit for agents and operators

Fieldwork Verification Kit v1

The agent finished. Make the handoff useful.

One download. Your own workflow.

Turn selected files and your recorded checks into a handoff the next person can inspect. Three reusable skills and local Python tools create a receipt, check file hashes and generate a clear Markdown evidence memo. You perform the task checks; the kit organizes the evidence.

  1. Choose the original requirements and collect the local deliverable.
  2. Record the checks and verify the artifact with the local tool.
  3. Generate a Markdown memo for the next person or agent.

Use it before handing off a report, code change, document or other local artifact.

Executed local example · macOS

From file to documented handoff.

A generated Markdown handoff was copied to a receiving folder, opened and checked. Fieldwork verified its evidence files and produced the memo you can read below.

  1. Open the receiving copy: 296 bytes read as UTF-8.
  2. Check the required sections: all three headings present.
  3. Verify identity: source and receiving copy have matching SHA-256 hashes.

Read the generated handoff memo → · Download the worked example

Open the example deliverable · Inspect the receipt

Actual checker output · excerpt

record_valid: true
artifact_hashes_checked: true
check_counts.passed: 3
all_checks_recorded_passed: true

Read the full JSON result

The included tool validates the receipt and file hashes; the recorded checks describe this local example.

Three jobs. One reusable kit.

Review a deliverable

Keep the original requirements beside the result. Validate the receipt and verify local artifact hashes.

python3 tools/fieldwork.py check examples/synthetic-pass.json --root examples/artifacts

Compare workflow runs

Compare records with matching contracts, inputs and environments. See recorded metric changes alongside the evidence.

python3 tools/fieldwork.py compare examples/synthetic-pass.json examples/synthetic-incomplete.json

Write the handoff memo

Render the evidence record as readable Markdown for the next person or agent.

python3 tools/fieldwork.py memo examples/synthetic-pass.json --root examples/artifacts

Commands use the included synthetic examples. Run them from the extracted kit folder; use your own receipts and evidence for real work.

Inside your download

Free or paid: choose the work you need

Free guides and Memo

Read a method and record your decision.

  • Eight task-specific guides and editable decision records
  • Record-consistency checker with optional guide and manifest comparison
  • Fictional worked sample and offline reader
  • Published Weekly issue 001 on background-job handoffs

Explore the free guides

Fieldwork Verification Kit

Build a repeatable evidence workflow.

  • Three reusable SKILL.md workflows
  • Receipt validation and optional local artifact SHA-256 checks
  • Compatible-receipt comparison with recorded metric deltas
  • Markdown memo renderer, Fieldwork method Memo 001, template, examples and tests

Use the free guides to choose a method and record a decision. The kit adds a reusable receipt, validation, comparison and memo workflow.

Read the free weekly Memo: When background jobs look finished before they are usable

Before you start

Does this replace Git or my handoff notes?

Keep your existing repository and notes. Fieldwork adds a structured receipt, selected-file hash checks and a generated evidence memo. Store those outputs alongside the work wherever you already keep it.

How is this different from the Comparison Lab?

Fieldwork keeps a deliverable, its checks and a handoff memo together, and compares two compatible receipts. Choose the $12 Workflow Comparison Lab when you want to plan repeated trials across tasks and summarize completion, time, effort and recorded cost across matched runs. You can use either pack on its own.

Explore the $12 Comparison Lab

Can my agent use it?

Give your agent the relevant SKILL.md file for the task. The files are readable instructions; global installation is optional. Automatic discovery and activation depend on the host.

What do I need?

A text reader for the skills and Memo; Python 3.10 or later for the local tools. Use the included local tools on macOS; read or adapt the portable skills in your existing agent workflow. On macOS, use real artifact-directory paths rather than symlink aliases such as /tmp or /var.

Does it send my files anywhere?

The included Python tool runs locally without networking, telemetry, an account or an API key.

Is it a subscription?

The kit is a one-time download with three skills, local tools and Fieldwork method Memo 001. The weekly Agent Optimization Memo is a separate subscription.

Is the included Memo the same as the free weekly issue?

The kit includes Fieldwork method Memo 001, “A file is not a finished workflow.” The free Weekly issue 001, “When background jobs look finished before they are usable,” is a separate publication. You can read that issue without buying the kit.

Can I use the free records directly in the paid tool?

Start with the kit’s included receipt template. It is designed for artifact checks, compatible-record comparisons and memo generation; the free guide records use their own format.

What does a successful check mean?

The tool validates receipt structure and can verify local file hashes. Its report also shows the task outcomes you recorded, giving you structure, artifact integrity and observations in one place.

Save the download and keep your browser cookies

Checkout sets a strictly necessary, signed recovery cookie before you leave this site. Return to your download page in the same browser, or preserve the same cookie jar in your agent, for up to 30 days after starting checkout. Your recovery cookie connects the purchase to your download.

Download recovery uses the same browser or cookie jar, rather than an email account. Save the ZIP to storage you control so you can keep using it on any device.

Files unlock after payment confirmation. For a pending payment, return in the same browser to check again. Recovery never requires another purchase. New downloads are limited to valid, non-refunded purchases.

License and purchase terms

Commercial license for your own and your organization’s work, including AI-agent and client projects. Share your own outputs; do not redistribute or resell the pack. Perpetual use of this edition. Prior MIT grants remain unchanged. See the full license for terms.

Read the full product license

Purchases are sold through Link. For non-delivery, goods that are defective, unfit for purpose or substantially misdescribed, or unauthorized or duplicate payments, request a refund through Link within 60 days. Link reviews eligibility case by case. Change-of-mind refunds are excluded except where consumer rights apply. EU and UK consumers have a 14-day cancellation period from delivery or access. Your statutory rights remain unchanged.

Link refund policy and request support · Sold through Link terms

One-time digital download

Make your next handoff easier to review.

Keep the deliverable, your recorded checks and a readable evidence memo together.

$12.00 USD plus applicable tax, shown at checkout. No recurring charge.

Checkout opens Stripe. Review the final total before payment. License and purchase terms