okhp3-custom-gpt-readiness
Skill3 starsChanged 2 months agoMITView on GitHub
Assess an incomplete or newly proposed Custom GPT concept for build readiness. Use when a user has a GPT idea, partial notes, pasted conversation, draft instructions, knowledge files, or an unfinished GPT and needs an evidence inventory, explicit gaps, targeted elicitation, a readiness score, or a structured handoff to okhp3-custom-gpt-builder.
In this repository57 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .agents/skills/vercel-react-best-practices/agents.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- architecture-decision-recordsSkill
- catalog-integritySkill
- frontend-designSkill
- okhp3-artifact-validationSkill
- okhp3-as-is-process-captureSkill
- okhp3-brand-style-registrySkill
- okhp3-capture-intakeSkill
- okhp3-chatgpt-project-migrationSkill
- okhp3-custom-gpt-builderSkill
- okhp3-decision-model-authoringSkill
- okhp3-elicitation-interviewsSkill
- okhp3-equilibrium-reviewSkill
- okhp3-evidence-standardSkill
- okhp3-future-state-strategySkill
- okhp3-gpt-skill-conversion-planSkill
- okhp3-graduation-gateSkill
- okhp3-handoff-packagingSkill
- okhp3-notion-capture-routerSkill
- okhp3-outcome-modeling-coreSkill
- okhp3-overkill-hill-brandSkill
- okhp3-process-captureSkill
- okhp3-process-gap-analysisSkill
- okhp3-process-intake-and-scopeSkill
- okhp3-project-promotionSkill
- okhp3-replit-build-in-publicSkill
- okhp3-replit-canvas-boardSkill
- okhp3-replit-design-pipelineSkill
- okhp3-replit-free-mode-autonomySkill
- okhp3-replit-github-syncSkill
- okhp3-replit-multi-artifactSkill
- okhp3-replit-repl-janitorSkill
- okhp3-replit-repository-janitorSkill
- okhp3-replit-task-executorSkill
- okhp3-repository-janitorSkill
- okhp3-repository-organizerSkill
- okhp3-session-handoffSkill
- okhp3-skill-catalogerSkill
- okhp3-skill-discoverySkill
- okhp3-skill-foundrySkill
- okhp3-source-backed-researchSkill
- okhp3-thread-context-extractionSkill
- okhp3-thread-extract-chatgptSkill
- okhp3-thread-extract-claudeSkill
- okhp3-triage-and-fileSkill
- okhp3-vite-github-pagesSkill
- production-build-verificationSkill
- published-keyboard-recovery-checkSkill
- publish-health-checkSkill
- publishing-trigger-checkSkill
- static-route-validationSkill
- validation-smokeSkill
- vercel-react-best-practicesSkill
- web-design-guidelinesSkill
- okhp3-skill-promotionSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 6 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 44
Produce a handoff: When the user is ready to continue, emit a builder-ready brief containing confirmed requirements, open questions, source …
- Distinctive: line 53
Targeted questions: only for gaps that change the build decision or acceptance tests.
- Distinctive: line 54
Phase and recovery map: when the source contains a chain or multi-step process.
- Very specific: line 55
Build handoff: when enough information exists, using the schema in references/readiness-dossier-schema.md.
- Distinctive: line 61
Every score has evidence or is explicitly marked missing.
- Distinctive: line 62
Every consequential claim has a claim class and a traceable source.
- Distinctive: line 63
Every blocker maps to a concrete question or artifact request.
- Distinctive: line 64
No duplicate questions for information already supplied.
- Very specific: line 72
Read references/readiness-dossier-schema.md before producing JSON or a machine-readable dossier.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
okhp3-custom-gpt-readiness
OverKill Hill P³ · overkillhill.com · github.com/OKHP3
Turn scattered GPT ideas and partial artifacts into a buildable, testable brief. This is an intake and gap-analysis skill, not the builder itself.
Scope
| In scope | Out of scope |
| Concept triage, evidence inventory, requirement elicitation, readiness scoring, and builder handoff | Writing production integrations, managing credentials, publishing a GPT, or claiming volatile platform limits |
Operating contract
- Inventory before asking. Inspect the supplied conversation, notes, files, prior prompts, prompt-chain stages, and existing GPT export. Record each artifact as
present,partial,missing, orconflicting; never ask for information already present. - Classify every claim. Quote or point to the source artifact for every
presentclaim. Label itverified_platform_fact,source_derived_practice,theory,preference,inferred, orunknown. A source assertion does not become a platform fact by being repeated. - Assess the eight readiness domains: job and audience, outcomes, boundaries, conversation contract, instruction behavior, knowledge/data, tools and permissions, and evaluation/governance.
- Identify blockers. A missing primary job, audience, safety boundary, allowed data source, or acceptance test is a build blocker. Do not declare a concept ready while a blocker remains unresolved.
- Ask only high-yield questions. Return the smallest question set that closes the largest blockers. Group questions by domain and explain why each answer matters.
- Score transparently. Score each domain 0 to 3:
0 missing,1 vague,2 usable with assumptions,3 explicit and evidenced. Report the total, percentage, blockers, assumptions, and confidence. A score is not a substitute for blocker review. - Choose a disposition:
ready_for_builder,ready_with_questions,needs_artifact_recovery,not_a_custom_gpt, orblocked_by_authority. Explain the decision. - Plan maturity from the start. For each acceptance test, state its evidence source, the failure it guards against, and how a future revision will be evaluated. For a multi-step build, produce an intake, contract, configuration, challenge, release, and recovery map. Each stage needs an input, observable exit gate, and recovery action. Treat untested claims as hypotheses, not release criteria.
- Produce a handoff. When the user is ready to continue, emit a builder-ready brief containing confirmed requirements, open questions, source evidence, non-goals, acceptance tests, safety constraints, and unresolved platform facts marked
verify.
Required output
Return these sections in order:
- Readiness verdict with disposition, score, confidence, and blockers.
- Evidence inventory table with
domain,status,claim_class,evidence,source, andowner/action. - Gap register separating blockers, important gaps, assumptions, conflicts, and unverified theories.
- Targeted questions only for gaps that change the build decision or acceptance tests.
- Phase and recovery map when the source contains a chain or multi-step process.
- Build handoff when enough information exists, using the schema in
references/readiness-dossier-schema.md.
Use synthetic or public-safe examples. Do not request secrets, tokens, private customer data, or employer-confidential material. If the concept is actually an Agent Skill, MCP server, fine-tuned model, or ordinary prompt, explain the mismatch and route to the appropriate workflow. Use custom-gpts/consolidated/EVIDENCE-REGISTER.md as the repository's interpretation standard when it is available.
Quality gates
- Every score has evidence or is explicitly marked missing.
- Every consequential claim has a claim class and a traceable source.
- Every blocker maps to a concrete question or artifact request.
- No duplicate questions for information already supplied.
- Acceptance tests are observable and testable, not aspirations.
- Current platform behavior is labeled
verified,unverified, orneeds verification. - The handoff is sufficient for
okhp3-custom-gpt-builderto begin without repeating intake. - A chain status or completion label is not treated as proof; each proposed phase has an observable gate and recovery path.
Handoff
Read references/readiness-dossier-schema.md before producing JSON or a machine-readable dossier. Use evals/evals.json as calibration for expected behavior. After a ready_for_builder verdict, load okhp3-custom-gpt-builder.
About
Built by Jamie Hill · OverKill Hill P³ Published at github.com/OKHP3 Part of the OKHP3/skillz Agent Skill library. MIT License -- free to use, fork, and adapt. A nod to the source is appreciated.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

