Agent Registry
Track each agent, model, tool set, best use, avoid list, and verification needed.
A $29 Notion and spreadsheet template pack for tracking agent runs, prompt versions, verification commands, cost notes, and repeat failure modes.
Checkout is not live yet. The order request records purchase intent and preferred checkout; revenue is counted only after checkout, receipt, or seller-dashboard evidence exists.
Track each agent, model, tool set, best use, avoid list, and verification needed.
Record every meaningful run with objective, inputs, changed files, evidence, and follow-up.
Version prompts with change reasons, known failure modes, and lightweight eval scores.
Keep exact commands and pass signals so "done" means something concrete.
Compare estimated tokens, direct costs, and human minutes saved by workflow.
Find prompt, model, verification, context, scope, and handoff leaks before they repeat.
Choose the agent, model, review path, fallback route, and proof gate before the run starts.
Compare old and new prompt behavior against scope, verification, stop, and handoff cases.
Audit objective, scope, proof, cost, handoff, security, and revenue claims before accepting agent output.
Mark agent work done only after objective, scope, behavior, regression, evidence, handoff, and revenue gates pass.
Accept or hold agent output with objective, scope, regression, security, handoff, and revenue proof.
Review secrets, personal data, auth, payment, dependency, public sharing, and prompt-leak risk.
Record timeline, impact, missed detection, recovery evidence, and prevention rules after agent failures.
Score run capture, scope control, verification, prompt control, cost control, failures, handoff, and revenue proof.
Choose Codex, Claude Code, or Cursor from proof required, handoff cost, local tools, IDE context, and stop conditions.
Convert repeat mistakes into detection rules and prevention rules.
Use the tour, ROI page, preview index, Gist, sample CSV files, and public templates to decide whether the full pack fits your agent workflow.
Use the order-intent page to prepare a clean request, acknowledge the $29 target price, and choose the checkout you prefer when a real seller channel is available.
The paid ZIP stays private. No order is counted as revenue until checkout, receipt, or seller dashboard evidence proves payment.
Compare the free preview with the paid pack: agent registry, run log, prompt library, verification ledger, cost tracker, failure modes, and weekly review.
The gallery groups run acceptance, cost control, handoff, safety, workflow, and buyer-proof templates into a search-friendly HTML page.
The keyword page routes Codex, Claude Code, Cursor, and Notion-template buyers to the sample dashboard, import builder, template gallery, and $29 request path.
The free maturity check rates run capture, scope control, verification, prompts, costs, failure handling, handoff, and revenue proof before a buyer requests access.
The free builder turns one AI-agent task into a scoped brief, acceptance log row, verification command, and order-ready proof packet.
The sample dashboard renders the run log, cost tracker, agent registry, prompt library, verification ledger, and failure modes so buyers can see the workspace shape before requesting access.
Use the interactive calculator to compare the $29 pack with avoided verification gaps, repeated prompt mistakes, scope drift, and weak handoffs.
Use the order-intent page to generate a copy-ready buyer note with tools, preferred checkout, purchase intent, and the revenue rule.
The pricing page shows what buyers get, checkout status, delivery proof, and the rule that demand signals are not counted as revenue.
Use the public cost tracker template to record agent, model, task, estimated tokens, direct cost, human minutes saved, and ROI notes.
Use the public audit to review prompt ambiguity, wrong model choice, weak verification, missing context, scope drift, repeated failure modes, and lost handoff.
Use the public routing matrix to choose agent, model, review path, escalation rule, fallback route, and verification proof before spending tokens or review time.
Use the public regression test to compare old and new prompt versions against objective clarity, scope control, verification, stop conditions, and handoff quality.
Use the public acceptance suite to check objective match, scope match, behavior proof, regression coverage, security/privacy risk, handoff quality, and revenue or delivery proof.
Use the public security checklist to review secrets, personal data, auth and permission risk, payment claims, public assets, dependencies, destructive actions, and prompt leakage.
Use the public incident report to capture timeline, impact, root cause, missed detection, recovery evidence, stakeholder follow-up, and durable prevention rules.
Use the public maturity scorecard to rate run capture, scope control, verification, prompt control, cost control, failure handling, handoff, and revenue proof.
Use the public picker to decide whether Codex, Claude Code, or Cursor should own implementation, review, UI QA, prompt work, release work, or handoff.
Public Gists mirror the compact run log, Codex workflow tracker, Claude Code workflow tracker, Cursor workflow tracker, scope freeze, revenue proof, Notion import guide, PR review scorecard, verification command library, paid delivery manifest, buyer decision matrix, checkout readiness FAQ, buyer intake packet, purchase approval brief, workflow stack comparison, cost leak audit, model routing matrix, prompt regression test, acceptance test suite, security review checklist, incident report, maturity scorecard, Codex vs Claude Code vs Cursor picker, verification ledger, and handoff template outside the launch repo, with early-access links back to the full pack.
Use the public registry template to track best use, avoid list, model, tools, prompt contract, verification, owner, and status for each agent.
Use the public ledger template to record exact commands, URLs, screenshots, release assets, and payment gates before an agent run is treated as complete.
Use the public prompt library template to track prompt text, use case, version, change reason, known failure, and eval score.
Use the public failure modes template to capture symptoms, root causes, detection checks, fixes, and prevention rules for future AI-agent briefs.
Use the public weekly review template to pick the best run, worst run, biggest time save, cost decision, memory update, and next revenue action.
Use the public release gate template to map build, tests, visual QA, secret scan, release asset, revenue claim, and rollback path to proof.
Use the public review brief template to define target, risk area, expected behavior, evidence, files, do-not-touch scope, finding format, and verification needed.
Use the public failure RCA template to capture expected result, actual result, evidence, root cause, missed detection, fix, prevention rule, and proof required before done.
Use the public run brief template to define objective, allowed scope, do-not-touch paths, required inputs, expected output, verification, and hard stops before an agent edits anything.
Use the public Codex tracker to record repository, goal, allowed files, inputs read, changed files, verification evidence, public proof, blocker, launch impact, and follow-up.
Use the public Claude Code tracker to capture mode constraints, allowed files, do-not-touch paths, inputs read, changes, verification evidence, review notes, blockers, and next action.
Use the public handoff template to preserve the original goal, current state, files touched, evidence, blockers, do-not-revert notes, next action, and completion requirement.
Use the public Cursor tracker to capture mode, allowed files, context provided, files changed, diff summary, verification evidence, manual review needed, blockers, and next action.
Use the public scope freeze template to define allowed files, do-not-touch paths, required inputs, allowed commands, forbidden commands, stop conditions, verification, and revenue claim rules.
Use the public revenue proof template to record price, revenue target, paid orders needed, checkout URL, seller dashboard, receipt evidence, gross revenue proven, public metrics, and signals that must not be counted.
Use the public Notion import guide to create the dashboard, import CSV databases, choose first views, preserve revenue proof rules, and avoid premature schema changes.
Use the public PR review scorecard to score scope match, correctness, tests, verification, regression risk, security/privacy risk, user-facing quality, and maintainability.
Use the public command library to map code, UI, release, secret, rollback, and revenue risks to exact commands, pass signals, fail signals, and evidence gaps.
Use the public delivery manifest to inspect delivered files, ZIP digest requirements, archive test proof, buyer acknowledgement, support boundary, and revenue counting rules.
Use the public matrix to compare current cleanup cost, expected saved time, pack surfaces, buy signals, do-not-buy signals, and preferred checkout before opening an early-access request.
Use the public FAQ to inspect common objections, checkout gates, delivery proof, and the rule that revenue requires checkout, receipt, payout, or seller-dashboard evidence.
Use the public intake packet to record buyer tools, recurring agent problem, cleanup cost, preferred checkout, receipt requirement, delivery path, and blocker without counting it as revenue.
Use the public approval brief to summarize price, business case, expected cleanup time saved, receipt requirement, delivery method, and the rule that approval is not revenue.
Use the public comparison to score run evidence, verification proof, prompt versioning, handoff, revenue proof, delivery proof, and setup speed before requesting access.
The preview index collects the landing page, tour, ROI calculator, templates, Gist mirror, release, metrics, and early-access path.
Use the public checklist before an agent run to define scope, stop conditions, verification, and handoff evidence.
Use the public run log template to capture objective, scope, changed files, exact verification, result, and follow-up.
Seven paid orders reach $203 gross. Checkout is not live yet, so the issue form captures purchase intent, preferred checkout, and delivery acknowledgement without publishing the paid ZIP.
A Notion-ready Markdown workspace, 15-minute quickstart, import guide, six CSV database imports, twenty-eight operating templates, weekly review, and launch examples for AI-agent workflows.
Solo builders and engineers using Codex, Claude Code, Cursor, or local AI coding agents who need a durable record of prompts, outputs, verification, costs, and repeated failures.
The product is packaged, but paid checkout needs an authenticated seller account. Until then, this page captures early-access demand without publishing the paid ZIP.
Yes. The public preview includes sample CSV files and launch copy. The full product pack contains the complete workspace, all starter databases, and operating templates.