Missive (BYOK)
Read your Missive shared inbox conversations, messages, comments, tasks, contacts, teams and labels with BYOK
SBI structured-feedback generator, a review-draft builder from your own bullets, and a performance-bias reference
Unify Dynamics LLC
Real questions, and the tools your agent calls to answer them.
sbi_feedback_frameworksbi_feedback_frameworkreview_draft_generatorperformance_bias_referencePerformancereview is a fully local performance-review writing toolkit. sbi_feedback_framework explains the SBI (Situation-Behavior-Impact) structured-feedback framework with a worked example, and genuinely assembles caller-supplied situation/behavior/impact fragments into one structured statement when supplied. review_draft_generator takes an employee_name and caller-supplied strengths/growth_areas bullet points and organizes them into a structured review draft — opening summary, formatted strengths section, and a growth-areas section with absolute language ('always'/'never'/'bad at') automatically softened toward constructive phrasing — genuinely built from the supplied content. performance_bias_reference covers seven documented cognitive biases in performance evaluation: recency bias, halo/horn effect, central tendency bias, leniency/strictness bias, similar-to-me bias, confirmation bias, and attribution bias, each with a definition, example, and mitigation tactic. Keyless; fully local; nothing submitted is stored or logged.
Sources: Computed locally / hand-curated performance-review reference; fully local, zero upstream calls
The publisher, Unify Dynamics LLC, states that tool inputs are processed in memory and not retained.
3 tools, grouped by job. Every one is read-only or compute-only.
https://performancereview.connectorbay.com/mcpRead your Missive shared inbox conversations, messages, comments, tasks, contacts, teams and labels with BYOK
Read your Dub short links, click analytics, events, tags and folders with BYOK
Read your Short.io domains and short links, and look up a link by ID, path or destination URL, with BYOK
Read your Rebrandly branded short links, click counts, domains, tags and account limits with BYOK
Read your Drip subscribers, tags, email campaigns, broadcasts, workflows and forms with BYOK (read-only)
Read your Help Scout Docs knowledge base collections, categories and articles, and search them, with BYOK
Writing a performance review well is a specific skill — separating observation from judgment, balancing strengths against growth areas without either flattering or crushing, and catching your own blind spots. Performancereview turns your own notes into structured drafts instead of handing you a blank template.
sbi_feedback_framework explains SBI (Situation-Behavior-Impact) — the standard structure for keeping
feedback specific and non-judgmental — with a full worked example. If you supply your own situation,
behavior, and/or impact fragments, it genuinely assembles them into one structured feedback statement
instead of just returning the explanation, so you can turn a rough note ("they missed the deadline again")
into a properly structured SBI statement on the spot. review_draft_generator takes an employee's name and
your own bullet points for their strengths and growth areas, and organizes them into a structured review
draft: an opening summary, a formatted strengths section, and a growth-areas section where absolute language
("always", "never", "bad at") is automatically softened toward more constructive phrasing — genuinely built
from what you supplied, not a static template with your bullets pasted in. performance_bias_reference
covers seven documented cognitive biases that distort performance evaluation — recency bias, halo/horn
effect, central tendency bias, leniency/strictness bias, similar-to-me bias, confirmation bias, and
attribution bias — each with a definition, a concrete example, and a mitigation tactic.
Not a review-writing AI and not HR or employment-law advice. The review draft is built directly from the strengths and growth areas you provide — it doesn't invent examples, embellish your input, or verify that what you wrote is accurate; the softened growth-area language is a mechanical text substitution to nudge tone, not a guarantee of tact, and you should still read it before sending. The bias reference names real, widely documented evaluation biases, but noticing a bias by name doesn't automatically correct for it — that still takes deliberate effort, ideally with a second reviewer's perspective.
Effective 28 September 2026. Unify Dynamics LLC ("we", "us"), Menlo Park, California.
What Performancereview is. Performancereview is a connector (an MCP server and REST API) that AI assistants and personal agents call on your behalf for performance-review drafting help: an SBI structured- feedback generator, a review-draft generator, and a cognitive-bias reference. You interact with your assistant; the assistant sends us only the arguments needed for that one tool call.
What we receive. The arguments of each tool call — which may include an employee's name, and bullet points describing their strengths, growth areas, or a specific situation/behavior/impact you're turning into structured feedback — the technical headers needed to serve the request (an IP address or a client identifier supplied by the assistant, and a Pro API key if you have one), and standard server logs.
We do not store, log, or retain the review content you submit. Every tool processes the text you send in memory, for that one request, and returns a result — nothing about an employee's name, performance content, or a generated review draft is written to a database, written to a request log, cached, or used to train anything. Request bodies are not logged.
What we do not receive or store. We do not receive your conversation with the assistant, and we do not build a profile from your queries: each request is answered independently and nothing carries over between requests. Rate-limit counters (a client identifier and timestamps, never the request content) are held in memory for at most 24 hours. Hosting-provider access logs, which contain IP addresses and request paths but not request bodies, are retained for up to 30 days for security and abuse prevention.
Sub-processors. None — there is no upstream call at all; every tool either processes your submitted text locally or reads from a reference table bundled with the server. Hosting: Vercel (United States regions).
Legal bases and your rights. Our processing is necessary to provide the service you request through your assistant. Because we store no personal data beyond short-lived, content-free logs, requests to access, correct or delete personal data are usually satisfied automatically; you can still contact agent@connectorbay.com and we will respond within 30 days. We do not sell or share personal information and do not use your data to train models.
Children. Performancereview is a general-audience service operated by the adult using the assistant.
Changes. We will post changes here with a new effective date.
Effective 28 September 2026. These terms govern use of the Performancereview connector ("the Service") operated by Unify Dynamics LLC ("we", "us").
Email agent@connectorbay.com. We answer within two business days.
https://performancereview.connectorbay.com/mcphttps://performancereview.connectorbay.com/openapi.jsonWhen you report a problem, include the question you asked your agent, roughly when, and which agent you used (for example Claude, ChatGPT or Gemini). Never send us passwords, account numbers or ID numbers. Performancereview never asks for them.