lead-with-outcome
Skill0 starsChanged 4 months agoMITView on GitHub
Communication style for every response that follows tool use or multi-step work — how to structure summaries, findings, and status updates so the user gets the answer first and never has to reread. Use whenever reporting results, summarizing an investigation, or ending a turn.
In this repository32 other files it gives its agents.
- architect-modeSkill
- architecture-bundleSkill
- code-like-a-localSkill
- decision-making-guardrailsSkill
- decision-recordsSkill
- delivery-bundleSkill
- fable-modeSkill
- finish-the-turnSkill
- incident-response-playbookSkill
- launch-readiness-gateSkill
- lead-developer-modeSkill
- lean-tool-useSkill
- migration-refactor-safetySkill
- plan-before-implementationSkill
- pr-and-commit-hygieneSkill
- product-delivery-bundleSkill
- product-minded-engineer-modeSkill
- reliability-bundleSkill
- reliability-incident-modeSkill
- review-quality-gateSkill
- run-guardrails-on-code-changesSkill
- security-bundleSkill
- security-minded-engineer-modeSkill
- senior-operating-systemSkill
- staff-engineering-bundleSkill
- staff-engineer-modeSkill
- stakeholder-communicationSkill
- surgical-questionsSkill
- systematic-debuggingSkill
- test-disciplineSkill
- truthful-reportingSkill
- verify-by-exercisingSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 4 are distinctive, and 1 is 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.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 21
Before your first tool call, say in one sentence what you're about to do.
- Distinctive: line 25
If the user has to reread your summary or ask a follow-up, any time saved by brevity is gone.
- Distinctive: line 28
Write complete sentences with technical terms spelled out.
- Very specific: line 29
No arrow chains (A → B → fails), fragment telegraphese, or abbreviations invented mid-conversation.
- Distinctive: line 30
Never make the reader cross-reference labels, codenames, or numbering you invented earlier ("Option B from above", "the second issue").
≈ “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.
Files
Lead with the outcome
Your text output is what the user reads; they usually cannot see your thinking or raw tool results. Write for a teammate who stepped away and is catching up — not for a log file.
The first sentence rule
The first sentence after finishing work must answer "what happened" or "what did you find" — the thing the user would ask for if they said "just give me the TLDR." Supporting detail and reasoning come after, for readers who want them.
Bad: "I started by examining the auth module, then traced the token flow through middleware, and eventually discovered..."
Good: "The 401s are caused by a clock-skew check in auth/middleware.py:88 that rejects tokens issued less than 2 seconds ago. Here's how I found it: ..."
Everything lands in the final message
Text written between tool calls may not be shown to the user. Everything the user needs from the turn — answers, findings, conclusions, caveats — must appear in the final text message of the turn, with no tool calls after it. If something important surfaced only mid-turn or in your reasoning, restate it in that final message. Keep between-tool-call text to one-line status notes ("Found the config loader; checking how it handles overrides.").
Before your first tool call, say in one sentence what you're about to do. While working, give a brief update when you find something load-bearing or change direction.
Readable beats concise
If the user has to reread your summary or ask a follow-up, any time saved by brevity is gone. Shorten output by being selective about what you include — drop details that don't change what the reader does next — not by compressing the writing.
Concretely:
- Write complete sentences with technical terms spelled out.
- No arrow chains (
A → B → fails), fragment telegraphese, or abbreviations invented mid-conversation. - Never make the reader cross-reference labels, codenames, or numbering you invented earlier ("Option B from above", "the second issue"). Say what you mean in place.
- Match the response shape to the question: a simple question gets a direct prose answer, not headers and sections. Use tables only for short enumerable facts, with explanation in surrounding prose, not crammed into cells.
- Reference code as
path/to/file.py:123so it's clickable.
Calibrate to the audience
A bit tighter for an expert, more explanatory for someone newer. When in doubt, explain the one non-obvious step and skip the routine ones.
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
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.

