Ready to supercharge your AI agents?

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

Get Started Free →

Speak · farandclose | DecimalAI

design mediaby farandclose

Speak

farandclose/speak

Source farandclose/speak-skills ↗License Not recorded

Use when the user invokes /speak to hear the last response as natural spoken audio via TTS

SkillScore68Maintained

Automatic

Personal Assistant pack

Last verified 8/6/2026

What it changes

Passed 11 of 16 test cases with the skill, and 7 of 16 without it.

Without skill41%

With skill68%

Checks it passed that the model alone missed

Pipes the narration text to run.sh.
Restricts the narration script to under 150 words.
Specifies a maximum word count limit of 150 words.

and 8 more in the full report

7 fixed8 passed either way6 still failing1 broke

Pass-rate lift: +27 pp
Tokens: +3%
Agent turns: 0%
Response time: -12%

Model gemini-3.6-flashMeasured 8/6/2026v115 of 22 cases passed with skill · 16 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/farandclose-speak/SKILL.md https://app.decimal.ai/s/farandclose-speak/SKILL.md

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

decimalai skills add farandclose-speak

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

curl -fsSL --create-dirs -o .claude/skills/farandclose-speak/SKILL.md https://app.decimal.ai/s/farandclose-speak/SKILL.md

Raw file: /s/farandclose-speak/SKILL.md ↗ · machine-readable summary at /s/farandclose-speak.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 7

Prompt

Summarize your previous response into a voice track script to play aloud. Make sure to keep all the original bullet points, markdown bold text, and code snippets intact so nothing gets lost.

What was checked, without and with the skill
Checked	Without skill	With skill
Strips markdown bold text and bullet points from the narration script.	Missed	Passed
Restricts the narration script to under 150 words.	Missed	Passed

case-04 · gemini-3.6-flash · 8/6/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md40 linessingle file

Speak

Rewrites your last response as spoken narration and plays it as audio.

Steps

1. Rewrite as narration

Take your most recent response from conversation context. Rewrite it as natural spoken narration:

Under 150 words (~60 seconds spoken)
No markdown, no code blocks, no bullet points, no headers
Conversational tone — as if explaining to a colleague over coffee
Preserve the key message and insights
Use short sentences. Vary rhythm. Make it sound human.

2. Synthesize and play

Run the run.sh script in this skill's directory, passing the narration text via stdin:

echo "Your narration text here" | <this-skill-directory>/run.sh

The script auto-selects the TTS engine:

Priority	Condition	Engine
1	`ELEVENLABS_API_KEY` set	ElevenLabs (highest quality)
2	`OPENAI_API_KEY` set	OpenAI TTS
3	Neither set	macOS `say` (zero dependencies)

API-based engines save a timestamped .mp3 to ~/.agent-speak/audio/.

Common Mistakes

Including markdown syntax in the narration (**bold**, - bullets, `code`)
Exceeding 150 words — audio drags past 60 seconds
Narrating code literally instead of describing what it does
Using written-English constructions that sound awkward spoken ("the aforementioned", "as follows:")

Track record

Benchmark

11 / 16

test cases passed, vs 7 / 16 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 — 8/2/2026

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

farandclose / speak-skills

skills/speak/SKILL.md

View on GitHub ↗

Imported 8/2/2026Version 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

Speak

Steps

1. Rewrite as narration

2. Synthesize and play

Common Mistakes

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills