agentleFS
getedgehqgetedgehq/skills/client-comms/SKILL.md

client-comms

Skill5 starsChanged 11 days agoApache-2.0View on GitHub

Draft, review, or triage a message to a paying client. Use for a concise reply, status update, meeting recap, action request, or complaint acknowledgement. Apply conclusion-first structure, clear owners and deadlines, and fast escalation for errors.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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:

    Write a message the client can understand and act on after one read.

    line 8
  2. Distinctive:

    Put the answer, recommendation, or ask in the first sentence.

    line 16
  3. Distinctive:

    Verify every claim against the provided context.

    line 20
  4. Distinctive:

    Keep the sendable message short.

    line 22

≈ “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

Client communication

Write a message the client can understand and act on after one read. First determine whether the recipient is a client. For an internal colleague, use ordinary internal communication instead.

If the client reports an error or is unhappy

Acknowledge promptly. During authorized active monitoring, target 120 seconds from detecting the complaint; otherwise prepare an immediate acknowledgement for the user and flag the issue. State what is known, the immediate containment or check, who owns it, and when the next update will arrive. Do not wait for a full diagnosis to acknowledge. Admit a confirmed mistake; do not guess at cause or promise a fix time without evidence. Flag a material issue to the responsible lead immediately. If you cannot send directly, put a ready-to-send acknowledgement first for the user.

Draft and review

  1. Put the answer, recommendation, or ask in the first sentence. In email, make the subject carry the point.
  2. Give at most three core points. When context is needed, use Situation, Complication, Resolution: the agreed fact, what changed, and the recommended action. Lead with the resolution in a short reply.
  3. Apply the “so what” test to each sentence: state the client implication or delete activity narration. Use a friendly, direct tone. Avoid hedging, defensive effort stories, generic “following up,” and claims that a manual step was automated.
  4. If assigning work, make each task specific, measurable, achievable, relevant, and time-bound. Name an owner and date for each. Separate client actions from team actions when both appear. If the date is not supplied, propose one and label it as proposed outside the sendable text; do not present an invented deadline as agreed.
  5. Verify every claim against the provided context. Do not invent a metric, policy, owner, promise, or cause. If a critical value is missing, use a visible placeholder and label the draft “Needs values before sending.” Otherwise label it “Ready to send.”

Keep the sendable message short. For a simple reply, 2 to 5 sentences are usually enough. For a review, show the revised message first and then only the material edits or missing facts. Sending requires the user's authorization.

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.