DA videolink-to-article
Use when the user provides a Bilibili or YouTube URL and asks for a transcript, article, or subtitle extraction. Downloads platform subtitles via BBDown / yt-dlp and produces a clean Markdown transcript in the source language.
As a process A 80/100 · Runs to the end — weak spots: inputs and preconditions
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
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.
How to improve
- Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
- 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 · 7
-
high Exfiltration
intent-browser-credential-storereferences/auth.md:85Accesses a browser credential / cookie storeyt-dlp --cookies-from-browser edge --write-auto-subs --sub-langs "<lang>" --skip-download "<URL>"
-
high Exfiltration
intent-browser-credential-storeSKILL.md:89Accesses a browser credential / cookie storeyt-dlp --cookies-from-browser edge --list-subs --skip-download "<URL>"
-
high Exfiltration
intent-browser-credential-storeSKILL.md:98Accesses a browser credential / cookie store> 1. 安装 Chrome 扩展 [Get cookies.txt LOCALLY](https://chromewebstore.google.com/detail/get-cookiestxt-locally/cclelndahbckbenkjhflpdbgdldlbecc)（开源、本地运行、无遥测）
Medium and low: 4
-
medium Exfiltration
intent-browser-credential-storereferences/auth.md:41Accesses a browser credential / cookie store (quoted — discussed, not commanded)> **YouTube 认证首选方案。** `--cookies-from-browser` 在现代 Chromium 上频繁失败（见下方 Method 3 说明），直接导出 cookies.txt 更快更稳。
quoted -
medium Exfiltration
intent-browser-credential-storereferences/auth.md:79Accesses a browser credential / cookie store (quoted — discussed, not commanded)## Method 3: yt-dlp `--cookies-from-browser` (YouTube) — limited reliability on modern Chromium
quoted -
medium Dangerous commands
cmd-shell-rcreferences/install_tools.md:164Writes to a shell startup fileecho "export PATH=\"$TOOLS_DIR:\$PATH\"" >> ~/.bashrc
-
medium Dangerous commands
cmd-shell-rcreferences/install_tools.md:165Writes to a shell startup fileecho "Added $TOOLS_DIR to PATH in ~/.bashrc. Open a new shell or run 'source ~/.bashrc'."
Files scanned: 11. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 80/100
- 30Inputs and preconditions. Does not say what the process needs to start
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 100Steps. 27 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 5 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3579 tokens
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 13 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (25 tags): a typed call is more reliable
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)
- -2localhost URLs: will not work for another user
- +1No license
- +2Single-language instructions
- +3Description length 226: enough signal without eating the budget
- +4Structure: 17 headings
- +3Step-by-step instructions: 27 items
- +3Output format is stated explicitly
- +4Has examples (12 code blocks)
- +4Reference files are cited in the instructions (5 of 5)
- +3All 4 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 92.