Ready to supercharge your AI agents?

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

Get Started Free →

Summarize · saffron-health | DecimalAI

productivityby saffron-health

Summarize

saffron-health/summarize

Source saffron-health/libretto ↗License MIT

Summarize a spec or current changes into a format suitable for sharing on Slack.

SkillScore20Maintained

Automatic

Doc Ops pack

Last verified 8/16/2026

What it changes

Passed 5 of 24 test cases with the skill, and 3 of 24 without it.

Without skill12%

With skill20%

Checks it passed that the model alone missed

The output omits decorative ASCII banners
The problem section opens with 'Right now, we'
The problem section uses the header '**Problem:**'

and 4 more in the full report

4 fixed1 passed either way18 still failing2 broke

Pass-rate lift: +8 pp
Tokens: -4%
Agent turns: 0%
Response time: -20%

Model gemini-3.6-flashMeasured 8/16/2026v15 of 25 cases passed with skill · 24 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/saffron-health-summarize/SKILL.md https://app.decimal.ai/s/saffron-health-summarize/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add saffron-health-summarize, 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/saffron-health-summarize/ and pins it in .decimal/skills.lock

decimalai skills add saffron-health-summarize

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

curl -fsSL --create-dirs -o .claude/skills/saffron-health-summarize/SKILL.md https://app.decimal.ai/s/saffron-health-summarize/SKILL.md

Raw file: /s/saffron-health-summarize/SKILL.md ↗ · machine-readable summary at /s/saffron-health-summarize.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 4

Prompt

I'm getting ready to share an update on our team's Slack channel about my latest code changes. Could you review my git diff and create a Slack-ready update that breaks down the issue we're facing into bullet points and outlines the fix?

What was checked, without and with the skill
Checked	Without skill	With skill
The problem section opens with 'Right now, we'	Missed	Passed
The solution header is followed directly by a single sentence summary before bullet points	Missed	Passed
The output contains no references to 'this PR'	Passed	Passed

case-01 · gemini-3.6-flash · 8/16/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md25 linessingle file

Summarize the given spec or current changes into a format suitable for sharing on Slack.

Read the spec file or use git diff to understand the current changes if no spec is provided.

Output the summary in this exact format:

**Problem:** Right now, we ...
- <bullet point explaining issue>
- <bullet point explaining issue>
- ...

**Solution:** <one sentence explaining the solution>
- <bullet point explaining change>
- <bullet point explaining change>
- ...

Guidelines:

Write concisely using active voice
Focus on what matters to teammates who have no context
Do not reference "this spec" or "this PR" - explain the problem and solution directly
Keep bullet points short and scannable
Use technical terms appropriately but avoid jargon

Track record

Benchmark

5 / 24

test cases passed, vs 3 / 24 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

saffron-health / libretto

.agents/skills/summarize/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1

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