PerfCopilot
HomeDocs
Documentation

Everything you need to write a fair review.

From your first connection to a sealed, delivered review. Set up PerfCopilot, connect your tools, and generate your first cited draft.

Free review generator

The free review generator drafts a performance-review paragraph from a role and a few accomplishments. No account, no signup, no credit card. It is a fast way to see the quality of a PerfCopilot draft and try the bias check, before you connect any of your own tools.

What it does

You give it two things and it returns three:

You provide:

  • A role (for example "Senior Software Engineer" or "Account Executive").
  • A few accomplishments, one per line, in plain language.

You get back:

  1. A drafted review of two to three short paragraphs, written in a fair, specific, evidence-oriented tone. It never invents facts beyond what you typed, and it never includes a numeric rating.
  2. A bias check. The same screen the full product uses scans the draft for loaded or biased language (recency, halo, gendered wording, and more) and lists what it found so you can adjust before you use the text.
  3. An evidence map. Each accomplishment is labeled with the source it would be cited from in the full product (GitHub pull requests, Jira issues, Slack threads, call recordings, docs and goals). This shows how the paid product grounds a review in your real work instead of asking you to remember and type everything.

How to use it

  1. Go to the free review generator.
  2. Enter the person's role.
  3. Add one accomplishment per line. Short, concrete phrases work best. For example:
    • "Shipped the billing v2 migration across 40 pull requests"
    • "Closed 12 Jira tickets ahead of schedule"
    • "Led the on-call rotation and two incident responses"
  4. Select Generate. The draft appears in a few seconds.
  5. Review the bias findings and the evidence map below the draft.
  6. Copy the text and edit it to fit your voice and your organization's format.

Tips for a better draft

  • Be specific. "Rebuilt the checkout flow and cut load time 40 percent" produces a sharper review than "did good work on the frontend."
  • Name the artifact. Mentioning PRs, tickets, calls, or docs both improves the draft and lights up the matching source in the evidence map.
  • Keep each line to one accomplishment. The generator reads them as a list; combining several into one line dilutes each.
  • Add three to six lines. One line is thin; a focused handful gives the draft enough to work with.

Limits

  • It is a short sample, not a full review. The free tool writes a brief draft to demonstrate quality. The full product produces longer, structured reviews with a rating per skill, peer feedback, and your organization's guidelines.
  • It does not save anything. Nothing you type is stored or tied to an account. Each request is independent.
  • It is rate limited. To keep the tool free and available to everyone, there is a limit on how many drafts one visitor can generate per hour. If you hit it, wait a bit and try again.
  • Evidence is illustrative. In the free tool the evidence map shows the category a claim would be cited from. It never fabricates a specific citation. The real citations come from your connected sources in the full product.

From the free tool to the full product

The free generator asks you to type accomplishments by hand. That is the one thing PerfCopilot removes: when you connect your tools, the product pulls the underlying work signals for each person automatically across a review cycle, aggregates them, and drafts a full review grounded in that evidence, with the same bias check applied.