simplify-report
Skill2 starsChanged 2 months agoMITView on GitHub
Explain supplied technical reports and already-produced tool output in clear everyday language while preserving the real outcome, severity, uncertainty, and actionable evidence. Use when the user asks what CI results, test summaries, audits, security findings, command output, status checks, or verification reports mean, including requests to simplify, decode, translate, summarize, or explain like they are five. Do not use for general prose editing, live diagnosis, or rerunning the underlying check.
In this repository18 other files it gives its agents.
- check-my-analysisSkill
- clean-handoffSkill
- clean-upSkill
- decision-coachSkill
- design-system-fitSkill
- feedback-to-findingsSkill
- form-flow-fixerSkill
- gameplanSkill
- knowledge-interviewSkill
- measure-successSkill
- meeting-briefSkill
- motion-resilienceSkill
- practice-coachSkill
- proposal-reviewerSkill
- reconcile-reportsSkill
- research-briefSkill
- ui-state-completenessSkill
- ui-stress-testSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
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
- Distinctive: line 12
Read only the supplied report or output.
- Distinctive: line 14
Check whether the evidence is incomplete, truncated, stale, or internally contradictory.
- Distinctive: line 21
Follow a format the user explicitly requests.
- Distinctive: line 23
Start with Bottom line. State the overall result and its practical meaning.
- Distinctive: line 29
Keep the full answer inside the bottom line and selected questions.
- Distinctive: line 33
Use calm, respectful, everyday language.
- Distinctive: line 39
Keep actionable evidence such as a count, severity, filename, command, or product term when it changes understanding or next steps.
- Distinctive: line 41
Do not assign blame or infer who caused a change.
- Distinctive: line 42
Do not invent remediation.
≈ “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
Simplify Report
Explain supplied technical output so the user can quickly understand what happened, why it matters, and what to do first. Preserve the report's actual conclusion.
Workflow
- Read only the supplied report or output. Do not rerun tools, checks, or commands unless the user separately asks. If no report or result is available, ask the user to provide it and stop.
- Identify the overall outcome, meaningful successes, failures or warnings, uncertainty, and the best supported first action. Do not infer an overall pass from successful subchecks.
- Check whether the evidence is incomplete, truncated, stale, or internally contradictory. State any limitation that changes what can safely be concluded.
- Separate reported facts from interpretation. Do not present a likely explanation as confirmed or invent a remedy the evidence does not support.
- Translate jargon into ordinary language. Preserve counts, severity, verdicts, and names the user needs to understand or act; omit internal identifiers that do not help.
- Draft the answer using the format below, then verify that it matches the report. Correct any softened failure, hidden uncertainty, unsupported claim, or misleading overall status before responding.
Default output
Follow a format the user explicitly requests. Otherwise:
- Start with Bottom line. State the overall result and its practical meaning.
- Include every relevant question below. Omit a question only when its answer would add no useful information.
- What worked? Explain the most meaningful successful parts without implying that they made the overall result pass.
- What needs attention? Explain the failure, warning, uncertainty, or unexpected result and why it matters.
- What should I do? Give the best first action supported by the report. When materially different choices exist, name the decision and briefly explain the options instead of choosing for the user.
- Use as many sentences as needed to make the explanation genuinely easy to understand. Prefer short sentences that explain one idea at a time, and stop when the user has enough context to understand the result and next step. Never omit a necessary explanation just to make the answer shorter.
- Keep the full answer inside the bottom line and selected questions. Do not add an opener, closing sentence, raw evidence dump, report path, taxonomy, or offer to explain more unless the user asks.
Writing rules
- Use calm, respectful, everyday language. "Explain like I'm five" means assuming no technical background, using simpler words, and explaining unfamiliar ideas without baby talk.
- Break complicated ideas into small steps. Use a familiar comparison only when it makes the meaning clearer.
- Lead with practical meaning, not the report format or the analysis process.
- Preserve severity. Never turn failed, blocked, incomplete, unsafe, or uncertain into passed or okay.
- Preserve mixed outcomes explicitly. State both a successful component and a failed overall result when both are true.
- Preserve uncertainty. If the evidence cannot establish something, say so plainly.
- Keep actionable evidence such as a count, severity, filename, command, or product term when it changes understanding or next steps.
- Distinguish a product failure from a check, tool, or environment failure.
- Do not assign blame or infer who caused a change.
- Do not invent remediation. If the report supports no safe next action, say what information or decision is needed first.
- For medical, legal, financial, privacy, or security findings, simplify the language without softening risks or removing necessary cautions.
Example
Input meaning: tests passed, two task files stayed in scope, one unrelated file changed, and the overall scope check failed.
Output:
Bottom line. The work passed its tests, but the overall scope check failed because an unrelated configuration file also changed.
What worked? The tests passed, and the two task files stayed in the agreed area.
What needs attention? config/prod.json changed outside that area. That extra change is why the overall check failed.
What should I do? Review the config change and decide whether to remove it or explicitly include it in the task.
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.

