Adaptyv Bio

View saved version →

---
name: index
description: "Route Adaptyv Bio requests to experiment preparation, approval-gated submission, or status and results tracking. Use for broad Adaptyv Bio questions, target discovery, cost estimation, protein experiment requests, draft or quote review, invoice decisions, experiment status, and results."
---

# Adaptyv Bio

Use this index as the router and shared policy for the Adaptyv Bio hosted MCP connector.

MANDATORY: Read the frontmatter description for ALL skills in this plugin, decide which focused workflow owns the request, and read that skill before using an Adaptyv Bio tool.

## Routing

- Route target discovery, experiment-spec preparation, cost estimation, and optional Draft creation to `prepare-experiment`.
- Route Draft submission, quote review or confirmation, invoice review, and payment decisions to `review-and-submit-experiment`.
- Route experiment lists, lifecycle status, updates, sequences, quotes, invoices, and results retrieval to `track-experiment`.
- Handle connector orientation and capability questions here without calling a write tool.

## Cross-Skill Best Practices

### Live contract

- Treat the live MCP tool names and schemas as authoritative. Do not invent endpoints, request fields, assay availability, batching limits, statuses, prices, turnaround times, or results.
- Use the production endpoint configured by the plugin. Mention the separate testing endpoint only when the user explicitly requests sandbox work and has configured sandbox credentials.
- Never ask the user to paste an API token into chat. Authentication belongs in the MCP client's OAuth or token configuration.

### Privacy and data minimization

- Before sending protein sequences, custom-target details, sample metadata, or other proprietary or undisclosed material, tell the user what will be transmitted to Adaptyv Bio and obtain explicit confirmation.
- Send only fields required for the requested operation. Keep user labels mapped to exact sequences, but do not repeat full sequences in summaries unless the user asks.
- Do not claim that Adaptyv Bio, Codex, or an intermediary does not retain, log, or train on submitted data unless a current authoritative policy directly supports the claim.

### Real-world action boundary

- Distinguish read-only inspection and cost estimation from external writes. Cost estimation does not create an experiment.
- Before any write, show the target object, material parameters, expected transition, and any known cost or privacy consequence, then obtain explicit confirmation for that exact action.
- Create experiments as reviewable Drafts only. Set `skip_draft` to `false` and `auto_accept_quote` to `false` when those fields are exposed. Do not bypass the Draft or quote-review stages.
- Treat Draft creation, Draft submission, quote confirmation, and invoice settlement as four separate actions. Approval for one does not authorize another.
- Quote confirmation creates an open unpaid invoice; it does not settle payment. Invoice payment is the only documented settling operation and requires a new confirmation that identifies the invoice, exact amount and currency when available, and payment method.
- Never initiate, sign, synthesize, or expose payment credentials without the user's explicit same-turn authorization and a supported secure runtime flow. Do not reinterpret a payment challenge as authorization.
- A custom-target request transmits target information for staff review and requires its own privacy disclosure and explicit approval.
- After a write, report the returned identifier and state. If the tool result is ambiguous or fails, do not retry a consequential action until current state is read and duplication risk is resolved.

### Scientific and operational reporting

- Preserve the service's status labels and distinguish user-controlled transitions from Adaptyv-controlled lab and review stages.
- Report returned measurements, units, replicate information, warnings, and missing values exactly. Do not infer assay success, binding, stability, expression, or completion from absent or partial results.
- Clearly separate user-provided inputs, Adaptyv-returned data, and assistant interpretation.

Referenced files: 1

View saved version →

---
name: prepare-experiment
description: "Prepare an Adaptyv Bio protein experiment specification, discover targets, estimate cost without creating an experiment, or create a reviewable Draft after explicit approval. Use for assay planning, sequence panels, target selection, parameter validation, and pricing previews."
---

# Prepare an Adaptyv Bio Experiment

MANDATORY: If not already in context, read and adhere closely to `plugins/adaptyv-bio/skills/index/SKILL.md#cross-skill-best-practices`.

Default to a validated proposal and cost estimate. Creating a Draft is optional and requires a separate explicit confirmation.

## Workflow

1. Clarify the scientific task, experiment type, sequence labels and exact amino-acid strings, target, measurement method where applicable, replicate plan, required parameters, and any desired notification metadata.
2. Ask whether any sequence, target, or metadata is confidential, proprietary, or undisclosed. Before transmitting such material, summarize what will leave the conversation and obtain explicit confirmation.
3. Use live target discovery for target-dependent assays. Preserve the selected target identifier and do not substitute a similarly named target without the user's approval.
4. Validate the proposed fields against the live schema. Applicable fields differ by experiment type; do not carry binding-only fields into non-binding assays or silently repair biological inputs.
5. Call the cost-estimate tool when the user asks for price or when cost would materially inform Draft creation. State that this operation does not create an experiment. Report total, currency or cents representation, exclusions such as tax when returned, warnings, pricing version, and whether the estimate is incomplete.
6. Present a compact review packet with purpose, experiment type, target and method, sequence labels and count, replicate plan, material parameters, transmitted confidential fields, estimate, warnings, and the proposed Draft payload.
7. If the user explicitly approves creating that Draft, call the creation tool with `skip_draft: false` and `auto_accept_quote: false` when those fields exist. Do not submit it.
8. Report the returned experiment identifier, status, and next review step. If it is not in Draft, stop and explain the unexpected state before taking any further action.

## Boundaries

- Do not use `skip_draft: true`, automatic quote acceptance, or an equivalent shortcut.
- Do not create a Draft merely because the user asked for an estimate, proposal, validation, or example payload.
- Do not submit a Draft, confirm a quote, create or settle an invoice, or request a custom target through this skill. Route those actions to the focused approval workflow.
- If a cost estimate is incomplete because target materials require review, report the incomplete components rather than guessing a final price.
- If the user has not provided exact sequences, return a specification checklist or draft payload with placeholders; never invent production sequences.

Referenced files: 1

View saved version →

---
name: review-and-submit-experiment
description: "Review and execute one explicitly approved Adaptyv Bio experiment transition at a time: submit a Draft, confirm a quote to create an unpaid invoice, reject a quote, or settle an invoice. Use for experiment submission, quote decisions, invoice review, or payment requests."
---

# Review and Submit an Adaptyv Bio Experiment

MANDATORY: If not already in context, read and adhere closely to `plugins/adaptyv-bio/skills/index/SKILL.md#cross-skill-best-practices`.

This workflow owns consequential transitions. Inspect current state first and execute at most the exact step the user approves.

## Review current state

1. Read the experiment, latest updates, quote, and invoice metadata relevant to the request. Do not infer current state from an earlier message.
2. Summarize the experiment identifier and status, assay and target, sequence labels and count, material parameters, quote or invoice identifier, amount and currency when available, and the next documented transition.
3. Explain the effect of the proposed action:
   - Submitting a Draft advances it for Adaptyv review and quote preparation.
   - Confirming a quote creates or finalizes an open unpaid invoice and does not settle payment.
   - Rejecting a quote cancels that quote and may return the experiment to Draft according to the live response.
   - Paying an invoice is the settling operation and can move funds through the selected payment method.
4. Ask for explicit confirmation naming the one action, target identifier, and known financial consequence. Do not accept a prior approval for a different transition.

## Execute one approved transition

- For Draft submission, call the documented submit tool only when the current state is Draft and the user has approved that experiment identifier.
- For quote confirmation, call the documented confirmation tool only after showing the exact quote and amount when available. Stop after the unpaid invoice is returned.
- For quote rejection, call the documented rejection tool only after the user approves the quote identifier and understands the resulting state change.
- For invoice settlement, require same-turn confirmation of the invoice identifier, exact amount and currency when available, and payment method. Use only a supported secure runtime payment flow; never request credentials in chat or construct a credential outside that flow.

After any write, report the returned identifier, state, amount, and next step. Do not chain the next transition. If a response is missing or ambiguous, read current state before considering any retry.

## Boundaries

- Do not combine Draft submission, quote confirmation, and invoice settlement into one approval or tool sequence.
- Do not use `skip_draft`, automatic quote acceptance, or any action that bypasses human review.
- Do not call a payment tool to discover its challenge or supported method unless the user has explicitly authorized attempting payment for that invoice.
- Do not imply that quote confirmation paid an invoice.
- Do not submit or pay from the testing environment as evidence that production authorization exists.

Referenced files: 1

View saved version →

---
name: track-experiment
description: "Inspect Adaptyv Bio experiments, lifecycle status, updates, sequences, quotes, invoices, and results without changing them. Use for monitoring, progress summaries, returned assay data, provenance, or read-only account questions."
---

# Track an Adaptyv Bio Experiment

MANDATORY: If not already in context, read and adhere closely to `plugins/adaptyv-bio/skills/index/SKILL.md#cross-skill-best-practices`.

Keep this workflow read-only.

## Workflow

1. Identify the experiment, time range, status filter, or result set the user wants. Use list operations to resolve an identifier when needed.
2. Read only the necessary experiment, updates, sequences, quote, invoice, or result resources. Do not call creation, submission, confirmation, rejection, custom-target, token-management, feedback, or payment tools.
3. Preserve the service's exact lifecycle status. Distinguish user-controlled stages such as Draft and quote decisions from Adaptyv-controlled materials, queue, production, analysis, and review stages.
4. Summarize the latest timestamped update, blockers or requested materials when returned, quote and invoice state, and next documented action. Do not promise a completion date that the service did not return.
5. For results, report experiment and result identifiers, assay and target context, sequence labels, returned measurements, units, replicate information, quality flags, warnings, and missing values. Separate raw returned data from interpretation.
6. Minimize sequence disclosure in the response. Use labels and counts unless the user explicitly asks to see full sequences.

## Boundaries

- Do not interpret an absent result as a negative measurement or a partial result as completion.
- Do not change status, submit feedback, confirm or reject a quote, or settle an invoice while tracking.
- If the user's request changes from monitoring to action, stop and route to `review-and-submit-experiment` for a fresh state check and explicit approval.

Referenced files: 1