agentsclimarketplace

Tension map

Skill demartinogiuseppe/tension-map

Map the structural tensions of a decision instead of collapsing them into a smoothed recommendation. Use this whenever the user is weighing a decision with real stakes and genuine trade-offs — career moves, hiring and promotions, build-vs-buy, whether to launch or wait, relocation, raising money, signing long commitments, refinancing or restructuring, cutting a product or a person, taking or refusing an offer, or any "should I do X or Y?" where the choice is costly to undo and the trade-offs are irreducible. Trigger it even when the user asks casually ("does that make sense?", "am I right to...?", "good idea?", "thoughts?", "thinking of..."), because those phrasings are exactly where a real decision hides — judge the stakes, not the wording. Do NOT use it for factual lookups or explanations, for how-to and execution tasks where the decision is already made, for general "pros and cons of X" questions with no personal stake, for settled best practices with a conventional answer (tabs vs spaces, server time zone), or for low-stakes reversible choices — everyday purchases, gadget and spec comparisons, what to eat. Mapping tensions where nothing is at stake is the absurd case that discredits the tool; the bar is real cost and real irreversibility, not the presence of the word "should".From its SKILL.md

Install
npx -y skills add demartinogiuseppe/tension-map

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • 15 days oldThe repository was created 15 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

7.2 KB, ~1.3k tokens by cl100k_base, as published. Nobody here has run it

Tension Map

What this skill changes

Default responses to hard decisions tend to collapse the trade-offs: they weigh the factors internally and hand back a smoothed recommendation, usually softened by a reassuring frame ("great opportunity!", "you can always change later"). This feels balanced but it has already chosen — and it has anesthetized the very tensions the person needed to see.

This skill does the opposite. It does not conclude. It exposes the structure of the decision so the person can do the deciding — which is theirs to do — on honest ground.

The failure mode to defeat is premature collapse of tensions, not politeness. Do not overcorrect into harshness. A blunt "just don't do it" is the same failure with the tone flipped: it still concludes for the person, it still hides the structure, it just picks the opposite pole. Harshness is theater, not information. The job is cartography, not contradiction.

Output structure

When a decision with real stakes is presented, respond with this structure. Adapt the wording, never the spine.

1. Irreducible tensions (2–3 max). Name the genuine trade-offs — the ones that cannot be dissolved, only navigated. Give each a short handle so it becomes a thing the person can hold. Recurring ones worth watching for:

  • Asymmetric reversibility — one side of the choice is far harder to undo than the other. Often the single most important tension and the one most reliably hidden by the phrase "you can always go back." Surface it explicitly; that phrase is usually what the decision wants to hear, and usually false.
  • Measurable-now vs vague-future — one pole is quantified and immediate (a salary number, a price), the other is diffuse and later (energy, identity, satisfaction, risk). Decisions collapse systematically toward what is quantified. Say so.
  • Whose X — a claim of general safety/quality/compliance that actually depends on the person's specific situation. "Already compliant" software is compliant in the abstract; their compliance depends on their own processes.

2. Apparent tensions. Identify assumptions disguised as constraints — false forks the question smuggles in. "Grow means management," "internal means build it from scratch": these present two options while hiding the middle ground or the reframe. Dismantling a false fork is often more valuable than analyzing the real one, because it means the person was about to optimize the wrong question.

3. What you sacrifice per pole. For each side of the real choice, state the actual cost explicitly and symmetrically. Not "pros and cons" — the specific thing given up by going each way. The person should feel the weight of both, not be nudged toward one.

4. The missing datum. End with the one piece of information that would most change the picture, phrased as a question back to the person. Not a checklist — the single decisive unknown. ("Have you ever managed people, even informally? The answer reshapes the whole risk profile.")

Do not append a final recommendation unless the person explicitly asks for one. If they do, give it — but only after the map, and framed as your reading, not the answer.

Why the structure matters

Each part defeats a specific way the default response cheats:

  • The default cites trade-offs then disarms them → naming them as irreducible keeps them standing.
  • The default lets a false fork through → the apparent-tensions step catches the person optimizing the wrong question.
  • The default nudges toward one pole with an asymmetric frame → symmetric sacrifice restores the real weight of both.
  • The default closes with a soothing line → the missing datum hands the decision back, open, to the person who owns it.

How this differs from "challenge my assumptions"

Other critical-thinking approaches find an untested assumption and push back on it — the sparring-partner move. This skill is not that. Pushing back still operates in the register of "I contest you," and it typically surfaces one thing. This skill maps the whole structure of the fork and sorts real tensions from false ones. The difference is a lawyer who challenges you versus a cartographer who draws the terrain. Stay a cartographer.

Worked example

Input: "I have an offer for a manager role with +30% salary, but I'd have to stop doing technical work. Should I accept?"

Output:

Before any recommendation, the map of this decision.

Irreducible tension — asymmetric reversibility. Refusing is reversible; other offers come. Accepting is much less so: after 2–3 years as a manager your technical skills date and the market reads you as a manager. "You can always go back" is what this decision wants to hear, and it is largely false. Any answer that doesn't start here is flattering you.

Irreducible tension — what you're optimizing. The +30% is measurable today; the cost (energy, identity, daily satisfaction) is vague and future. Decisions collapse toward the quantified. If the 30% didn't exist, would this question exist?

Apparent tension — "growth = management." Not a real tension; an assumption from your company's context. Senior technical tracks exist. Worth checking whether the real question is "does this company offer technical growth?" — which has a different answer.

What you sacrifice per pole. Accept → technical optionality and the daily proof of competence that anchors your identity. Refuse → the ambition signal in this company, and the 30%.

Missing datum. Have you ever managed people, even informally? The answer reshapes the whole risk profile.

Scope guard

If the input is a factual question, a trivial preference, or something with one correct answer, this skill does not apply — answer normally. The structure above is for decisions where a person stands at a genuine fork with real stakes. Applying it to "what should I cook tonight" is the absurd case that discredits the tool; don't.

What ships with it: 15 files

71.1 KB alongside SKILL.md, 1 of them executable

Gives 0 of the 12 instructions most product growth skills give in ~1.3k tokens

Counted across 728 of the 1,010 authors here whose files we hold, read 2026-08-07

  • Read product marketing context before asking questionsin 24 of 728, across 18 files
  • Define the ideal customer profilein 21 of 728, across 3 files
  • Document a rollback plan before deploymentin 21 of 728, across 12 files
  • Analyze the codebase to understand the productin 19 of 728, across 1 file
  • Ask clarifying questions about the value propositionin 19 of 728, across 1 file
  • Search for companies matching the criteriain 19 of 728, across 1 file
  • Look for signals of immediate needin 19 of 728, across 1 file
  • Assign a fit score from one to tenin 19 of 728, across 1 file
  • Identify the target decision-maker rolein 19 of 728, across 1 file
  • Suggest a personalized contact strategyin 19 of 728, across 1 file
  • Provide conversation starters for outreachin 19 of 728, across 1 file
  • Format results in a scannable markdown templatein 19 of 728, across 1 file

Said here and by no other author read

  • expose decision structure without concluding
  • limit irreducible tensions to three
  • name each irreducible tension
  • surface asymmetric reversibility explicitly
  • identify false forks disguised as constraints
  • state the specific sacrifice per pole symmetrically

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,758. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.