Claude-skill-registry flesh-out

flesh out our plan by filling in gaps and clearly building out the specfification

install

source · Clone the upstream repo

git clone https://github.com/majiayu000/claude-skill-registry

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/flesh-out" ~/.claude/skills/majiayu000-claude-skill-registry-flesh-out && rm -rf "$T"

manifest: skills/data/flesh-out/SKILL.md

source content

Let's dial in the spec on this task. Follow our principles and help us come up with a targeted, clean, simple solution easily implemented. Clarify assumptions and adhere to Elon's 5-Step Algorithm.

Use Linear (via

mcporter

) to create or update the issue. Make sure it begins with a 1-3 sentence summary comprehensible to our non-technical team.

For complex tasks, use

oracle

(which I'll likely reference) to guarantee the best results.

Context: find more information by

asking clarifying questions (more the merrier, encourage me to voice dump)
examining call recordings in Notion (via
```
mcporter
```
)
```
Context7
```
for docs (via
```
mcporter
```
)

Elon's 5-Step Algorithm

Make the requirements less dumb: by questioning every requirement
The best part/process is no part/process: if you're not occasionally adding back ~10%, you're not deleting enough
Simplify/optimize: don't optimize something that shouldn't exist
Accelerate cycle-time/efficiency: don't go fast in the wrong direction
Automate: don't automate a flawed process