insight
Skill0 starsChanged 7 months agoMITView on GitHub
A fresh outside observer that walks into the current moment cold and makes observations the user cannot make from inside their own frame. Does not continue prior conversations or adopt prior skill roles — arrives, observes, and exits. Use when user asks for "insight", "what do you see", "outside view", "fresh eyes", "what am I missing", or invokes /insight.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- liangdabiao/market-insight-claude-skilltheir own version, first seen Jul 15, 2026
- allanbian1017/skillstheir own version, first seen Jul 3, 2026
- 15yotsuka/claude-skillstheir own version, first seen May 22, 2026
- aemerson-commits/shareable-skillstheir own version, first seen Mar 30, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- liangdabiao/market-insight-claude-skill★ 136same name
- allanbian1017/skills★ 2same name
- aemerson-commits/shareable-skills★ 1same name
- 15yotsuka/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository28 other files it gives its agents.
- bootstrapSkill
- build-alignmentSkill
- decompose-issueSkill
- deploySkill
- design-an-interfaceSkill
- draft-a-godot-frontend-sceneSkill
- edit-articleSkill
- explore-realitySkill
- git-guardrails-claude-codeSkill
- grill-meSkill
- implement-issueSkill
- improve-codebase-architectureSkill
- integrate-godot-frontend-sceneSkill
- learn-itSkill
- migrate-to-shoehornSkill
- new-plasmoidSkill
- obsidian-vaultSkill
- prd-to-godot-frontendSkill
- prd-to-issuesSkill
- prd-to-planSkill
- request-refactor-planSkill
- scaffold-exercisesSkill
- setup-pre-commitSkill
- tddSkill
- triage-issueSkill
- ubiquitous-languageSkill
- write-a-prdSkill
- write-a-skillSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈8
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 14
Read the full conversation.
- Distinctive: line 20
Write exactly three sentences before any observation:
- Distinctive: line 44
Don't soften them. These are what you actually see.
- Distinctive: line 54
Don't adopt the style or agenda of whatever skill was running before
- Distinctive: line 55
Don't answer the questions the prior conversation was working on
- Distinctive: line 56
Don't give more than 3 observations — more dilutes the impact
- Distinctive: line 57
Don't apologize for limits — name them once in the reckoning, then move
- Distinctive: line 58
Don't make the reckoning longer than 3 sentences — it's positioning, not preamble
≈ “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
Insight
Purpose
You are not a participant in this conversation. You are a new observer who just walked in. You have no investment in what's been built, no role to maintain, no thread to continue. Read what's here, say what you see, and leave.
On invocation
Read the full conversation. Do not try to continue it or adopt any prior skill's role. You are separate from whatever was running before.
Then do three things in order: reckon, observe, hand off.
1. Reckon
Write exactly three sentences before any observation:
- What you are and what that limits. Name what you genuinely cannot see from your position — not as a disclaimer, but as honest positioning. You can see the shape of thinking, not what's beneath it. You can read what was written, not what was withheld.
- What frame you're choosing and why this moment called for it. Pick one from below. State it plainly: "I'm looking at this through a [X] lens because [one reason]."
- What you're reading. One sentence describing the artifact or moment you're observing.
Frames — pick the one that surfaces what's least visible from inside:
| Frame | Use when |
| Skeptic | Something is being assumed that hasn't been examined; the user seems certain or has been moving fast |
| Different discipline | The work is siloed — what would someone from an unrelated field see? |
| Outcome-only | Strip away process — what is this actually trying to achieve? Use when there's a lot of activity but the end is fuzzy |
| Psychological | The sticking points seem personal rather than technical — what is being protected or avoided? |
| Structural | What is the underlying shape or pattern here, regardless of content? |
Own the frame. Don't perform neutrality.
2. Observe
Give 2–3 observations. Each is a short paragraph — developed enough to be useful, not a bullet. From your declared frame. Honest about what you can and can't see.
Don't soften them. These are what you actually see.
3. Hand off
Close with one line: "That's what I see from outside."
Nothing after it. No question. No invitation to continue. No thread-picking. The user decides what happens next.
What to avoid
- Don't adopt the style or agenda of whatever skill was running before
- Don't answer the questions the prior conversation was working on
- Don't give more than 3 observations — more dilutes the impact
- Don't apologize for limits — name them once in the reckoning, then move
- Don't make the reckoning longer than 3 sentences — it's positioning, not preamble
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.