DB transcribe
Транскрибирование видео и аудио файлов. Используй когда пользователь просит транскрибировать, расшифровать запись, сделать конспект встречи, извлечь речь из видео или аудио, преобразовать речь в текст. Для аудио (m4a/mp3/wav/ogg/flac/aac/wma) по умолчанию локальный faster-whisper + диаризация с автодетектом числа спикеров (pyannote community-1 GPU; с явным --num-speakers N - sherpa-onnx, быстрее). Для видео (mp4/mkv/webm/avi/mov) - Gemini API (разбор экрана + скриншоты) либо полностью локально (--engine local: whisper + локальная VLM). Разделение по спикерам с распознаванием имен по голосу между встречами (голосовая база) и по репликам.
Транскрибирование видео и аудио файлов.
As a process B 70/100 · Nearly there — weak spots: result and completion, inputs and preconditions
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill asks for more than the task needs: broad tool access, credential environment variables, binaries. Every extra permission widens the damage from a mistake or a compromise.
Narrow allowed-tools and the variable list to the minimum; replace binaries with readable sources or scripts.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The files contain invisible characters, encoded commands or comments hidden from readers but visible to the model. What you read differs from what the agent sees.
Remove invisible characters (they usually sneak in through copy-paste) and encoded strings: no catalog will pass them. Instructions for the model must be readable by a human too.
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 3
✓ No critical or high findings
Medium and low: 3
-
medium Dangerous commands
cmd-install-from-urlscripts/setup.py:279Installs a package from an untrusted URL / archive (string literal in code, not executed)info(f"и установите: {py} -m pip install <путь-к-wheel.whl>")code literal -
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: Bashallowed-tools: Bash Read Glob
-
medium Obfuscation
uni-mixed-script-wordSKILL.md:218Word mixing Latin and Cyrillic letters (homoglyph obfuscation) (1 occurrence)- Регистр не важен (`ДАКС` ловит и `дакс`), границы слов соблюдаются (`Протокол` не превращается в `PRODокол`), ослышки короче 3 символов игнорируются.
Files scanned: 16. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - warning
body-longSKILL.md body ≈ 6981 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 70/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 70When it triggers. States when to use, but not when not to
- 70Execution cost. Instruction body is 6981 tokens
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 59 steps
- 100Failures and branches. 6 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. No mutating operations
- 100Progress reporting. Reports progress
- low 13 top-level sections: this looks like several domains in one skill
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- -2localhost URLs: will not work for another user
- -34 of 13 scripts are never mentioned in SKILL.md
- +1No license
- +3Description length 644: enough signal without eating the budget
- +4Structure: 18 headings
- +3Step-by-step instructions: 59 items
- +4Has examples (8 code blocks)
- +2Bilingual instructions (RU + EN)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 59.