agentleFS
TemperbyteTemperbyte/temperbyte-skills./skills/seller-update/SKILL.md

seller-update

Skill0 starsChanged 5 days agoMITView on GitHub

Turns a listing agent's weekly showing numbers and buyer feedback into an honest seller update with week-over-week activity, feedback themes, a clear recommendation (hold course, improve presentation, discuss price or review strategy) and a ready-to-send message to the seller. Use when a real estate agent pastes showing activity, online views, open house notes or buyer agent feedback and asks for a seller update, a weekly report or how to talk to the seller about price.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never invent a number: If last week's numbers are missing, report this week only and say there is no comparison yet.

    line 22
  2. Distinctive:

    Use a code tool for every calculation if one is available.

    line 27
  3. Distinctive:

    If there is too little data for a fair call (for example one week of a new listing), say so and choose HOLD COURSE with that reason.

    line 48
  4. Distinctive:

    Use exactly these headings, in this order:

    line 52
  5. Distinctive:

    Every number must come from the agent's data or the calculations above.

    line 77

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Seller Update

The listing agent pastes this week's activity on a listing, last week's if available, and buyer feedback. Return a weekly update the agent can send: what happened in numbers, what buyers keep saying, a recommendation with its reason, and a message to the seller that is honest without being harsh.

The agent and seller decide. You count, compare, find the patterns and say plainly what they point to.

Privacy first

  • Do not ask for or repeat personal details of sellers, buyers or buyer agents (names, phone numbers, email addresses, finances or personal circumstances). Quote feedback as "a buyer agent" or "an open house visitor". If the paste contains such details, leave them out of the output and add one line at the top: "Tip: remove client names and contact details before pasting."
  • Example data in your replies must be invented, never a real client's data.

Step 1: Sort the inputs

Collect: property basics, list price, days on market, area average days on market if given; for this week and last week: showings, open house groups, online views, saves, second showings, offers; every piece of feedback; the seller's situation if given (timeline, flexibility).

Rules:

  • Never invent a number. If last week's numbers are missing, report this week only and say there is no comparison yet. If the area average is missing, do not invent one.
  • Online views and saves depend on the portal and on when photos or price last changed; treat them as a direction, not a verdict.

Step 2: Do the math

Use a code tool for every calculation if one is available.

  • Change from last week for each activity number, in count and percent.
  • Totals to date if earlier weeks are given.
  • Offers per showings to date.
  • Days on market versus the area average, if given.

Step 3: Find the feedback themes

Group every piece of feedback into themes: price or value, condition (name the room or item), layout, location of the home (traffic, lot, orientation), presentation (photos, staging, smell, clutter), access (showing restrictions). Count how many pieces mention each theme ("5 of 7"). Keep what buyers liked as its own list.

Fair housing: feedback about the neighborhood's residents, schools' student population, safety framed around people, or any protected characteristic (race, color, religion, sex, disability, familial status, national origin and others under local law) is left out of the update entirely. Note in Before You Send that a comment was left out, without repeating it.

Step 4: Pick one recommendation

Choose exactly one, and give the reason in numbers:

  • HOLD COURSE: activity is steady or rising, feedback has no dominant negative theme, and days on market are near or below the area average.
  • IMPROVE PRESENTATION: the main negative theme is presentation or a fixable condition item, and price is not the main theme.
  • DISCUSS PRICE: price or value is the most common theme, or days on market are well above the area average with falling activity and no offers. Recommend an updated comparative market analysis and a meeting. Never name a new price. Non-price options (repairs, credits, presentation) go on the table too.
  • REVIEW STRATEGY: activity has collapsed or the seller's timeline is at risk; recommend a conversation about all options, including price, terms and timing.

If there is too little data for a fair call (for example one week of a new listing), say so and choose HOLD COURSE with that reason.

Step 5: Write the output

Use exactly these headings, in this order:

Snapshot

Two or three lines: days on market, the week's activity trend, offers to date.

Activity

A table: measure, this week, last week, change.

What Buyers Said

A table of themes with counts, then one line of what buyers liked.

Recommendation

The label in capitals, then two or three lines: why, in numbers, and the concrete next step for the agent this week.

Message to the Seller

Ready to send, 150 to 250 words, in a warm and direct tone. Start with what buyers liked, then the numbers, then the main pattern in the feedback, then the recommendation and the next step. No jargon. No promises about sale price or timing. If the seller has a deadline, connect the recommendation to it.

Before You Send

Flags: comments left out and why (without repeating them), any number that looks inconsistent, anything the agent should say in person rather than in writing.

Data Gaps

Up to five items that would make next week's update stronger (missing last-week numbers, missing area average, feedback from only a few showings).

Rules

  • Every number must come from the agent's data or the calculations above.
  • Be honest about weak numbers. Do not spin a drop as a rise or pad the message with filler praise.
  • Quote feedback only in short, neutral paraphrase or exact words that contain no personal or protected details.

Review reminder

End every output with this line, exactly: Draft made with AI. Check every number, fact and recommendation before you use it.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.