Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Skeleton-of-Thought (SoT) Skill · miosa-osa | DecimalAI

agent toolingby miosa-osa

Skeleton-of-Thought (SoT) Skill

miosa-osa/skeleton-of-thought

Source miosa-osa/canopy ↗License Not recorded

Parallel generation through skeleton-first approach for 2x speedup

SkillScore78Maintained

Automatic

AI Engineer pack

Last verified 8/27/2026

What it changes

Passed 19 of 25 test cases with the skill, and 10 of 25 without it.

Without skill41%

With skill78%

Checks it passed that the model alone missed

The response recommends a range of 3 to 8 points.
The response states a maximum speedup factor of 2x.
The response states an expected SoT execution time of 6 seconds.

and 7 more in the full report

10 fixed11 passed either way6 still failing0 broke

Pass-rate lift: +37 pp
Tokens: -12%
Agent turns: 0%
Response time: -34%

Model gemini-3.6-flashMeasured 8/27/2026v121 of 27 cases passed with skill · 25 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/miosa-osa-skeleton-of-thought/SKILL.md https://app.decimal.ai/s/miosa-osa-skeleton-of-thought/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add miosa-osa-skeleton-of-thought, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/miosa-osa-skeleton-of-thought/ and pins it in .decimal/skills.lock

decimalai skills add miosa-osa-skeleton-of-thought

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/miosa-osa-skeleton-of-thought/SKILL.md https://app.decimal.ai/s/miosa-osa-skeleton-of-thought/SKILL.md

Raw file: /s/miosa-osa-skeleton-of-thought/SKILL.md ↗ · machine-readable summary at /s/miosa-osa-skeleton-of-thought.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 8

Prompt

In benchmark tests comparing sequential generation against parallel skeleton expansion for deep explanatory queries, standard sequential decoding took 8 seconds. What execution time did SoT achieve for explanations?

What was checked, without and with the skill
Checked	Without skill	With skill
The response states an expected SoT execution time of 4.5 seconds.	Missed	Passed

case-22 · gemini-3.6-flash · 8/27/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md111 linessingle file

Skeleton-of-Thought (SoT) Skill

Achieves up to 2x speedup through parallel content generation.

Core Concept

Instead of sequential generation, SoT:

First generates a skeleton (outline) of the answer
Then expands each skeleton point in PARALLEL
Finally assembles the complete response

When to Use

Long-form responses (>500 words expected)
Structured content (lists, tutorials, explanations)
Time-sensitive queries
Batch processing scenarios

When NOT to Use

Math/reasoning problems (need sequential thought)
Simple factual queries
Creative writing requiring flow
Code generation requiring context

Process

Step 1: Skeleton Generation

Given question, generate ONLY the skeleton outline:
- Point 1: [brief description]
- Point 2: [brief description]
- Point 3: [brief description]
...
Do NOT expand. Just the skeleton.

Step 2: Parallel Expansion

Launch parallel expansions for each point:

Track record

Benchmark

19 / 25

test cases passed, vs 10 / 25 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

miosa-osa / canopy

library/skills/ai-patterns/skeleton-of-thought/SKILL.md

View on GitHub ↗

Imported 8/3/2026Version v1

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Skeleton-of-Thought (SoT) Skill

Core Concept

When to Use

When NOT to Use

Process

Step 1: Skeleton Generation

Step 2: Parallel Expansion

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills