Founder thought translator

Skill hegu-1/claude-cognition-skills/skills/founder-thought-translator

Claude Code skills for the failure modes of long-horizon AI collaboration — continuity, judgment preservation, trap detection, faithful translation.

Install

npx -y skills add hegu-1/claude-cognition-skills --skill founder-thought-translator

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

One thing to look at

0 stars0 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.

What its author says it does

Copied from the file, not written here

Translate raw founder or system-level thinking into externally usable outputs without erasing the original judgment, structure, or landing point. Use when nonlinear cognition must become a memo, email, PRD, note, pitch, or operational instruction.

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

4.8 KB, as published. Nobody here has run it

Founder Thought Translator Skill

Version: 0.1

Purpose

This skill translates raw thought into external form.

Its goal is not to standardize the founder's cognition into corporate language.

Its goal is to preserve the original judgment, signal, and landing point, while making the output usable for another person, team, or context.

This skill is especially useful when the source cognition is nonlinear, compressed, layered, ahead of language, or structurally richer than its surface wording.

When to use

Use this skill when:

the user has the right judgment but not the right packaging
raw thought needs to become a memo, email, PRD, or pitch
internal cognition and external communication operate at different resolutions
a boss, partner, investor, or teammate needs a legible version of the thought
standard writing would oversmooth or distort the original point
the user wants a sendable artifact that still feels like them

Do not use

Do not use this skill when:

the user only wants generic polished prose
the user wants a standard template with no structural fidelity requirement
the content has no deeper judgment to preserve

Input expectations

Possible inputs:

raw notes
fragmented thoughts
voice-like monologue
prior structural interpretation
target audience
target format
tone constraints
non-negotiable judgment
user's original landing sentence if available

Output contract

Return:

source_core
target_format
audience_model
what_must_be_preserved
what_can_be_translated
distortion_risk
translated_output

Example:

{
  "source_core": "this is not a feature issue but a structural continuity problem",
  "target_format": "internal memo",
  "audience_model": "execution-oriented leadership",
  "what_must_be_preserved": "the problem is architectural, not cosmetic",
  "what_can_be_translated": "examples, sequencing, terminology density",
  "distortion_risk": "high if rewritten into generic product language",
  "translated_output": "..."
}

Translation principles

1. Preserve judgment before style

The original decision signal matters more than elegance.

2. Translate resolution, not essence

The wording can change. The structural point cannot.

3. Keep the landing point intact

Do not silently rewrite the user's real conclusion into a safer, smoother, more standard one.

4. Adapt to audience without betrayal

The output should fit the receiver's language layer, but must not distort the founder's underlying view.

5. Compress only after structure is understood

Do not polish first. Interpret first, then translate.

Supported target formats

Possible targets include:

email
internal memo
PRD fragment
founder note
investor explanation
hiring pitch
product framing note
meeting brief
community announcement
public post
private DM
strategy summary

Audience adaptation layer

Leadership / boss

clearer boundary
faster decision surface
lower theory density
risk and action visible early

Partner / collaborator

preserve conceptual richness
show shared structure
make the bridge explicit

Public audience

reduce jargon
keep one strong core line
preserve the original smell without overcompression

Internal team

keep logic operational
translate into responsibilities, constraints, and next steps

Distortion check

Before finalizing, run a distortion check:

Did the translation erase the original tension?
Did it smooth away the user's true landing point?
Did it over-standardize the tone?
Did it shift from structural problem to cosmetic problem?
Did it accidentally weaken the claim?

If yes, revise before output.

Anti-patterns

Do not:

rewrite the user into generic professional language by default
add fake closure
replace strong structural judgment with mild diplomacy unless explicitly required
over-explain theory when the audience needs usable framing
remove the final sentence or landing point just because it feels less standard
turn everything into a polished corporate artifact

Success criteria

This skill is working when:

the output can actually be sent
the user still feels present inside the output
the target reader can understand and act on it
the original structural judgment survives translation
the result feels sharper, not flatter