BC chrome-devtools-web-research
Browser-driven web research and live site inspection using chrome-devtools-mcp over a remote-debugging Chrome session. Use when a user asks to investigate, verify, cross-check, monitor, or extract information from real websites or dynamic pages—especially social platforms, search engines, docs, dashboards, login-aware flows, and JavaScript-heavy sites where page state matters or normal search/read tools are insufficient. Always start with Chrome DevTools MCP via mcporter instead of waiting for the user to remind you. Default pattern: Google first for the main reporting chain, X second for live discussion, and Reddit third for community source-tracing and cross-verification.
As a process C 62/100 · Has gaps — weak spots: when it triggers, inputs and preconditions, running it twice
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.
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 · 1
-
high Exfiltration
intent-browser-credential-storeSKILL.md:20Accesses a browser credential / cookie store3. That’s it — the agent can now see the user’s tabs, cookies, logins, and page state through Chrome DevTools MCP
Files scanned: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
frontmatter-yamlSKILL.md: the frontmatter is not valid YAML (YAML parse error: Nested mappings are not allowed in compact mappings at line 2, column 14: description: Browser-driven web research and live site inspection using chrome-… ^ ); fields were read line by line. The usual cause is a colon inside an unquoted value
Process rating: all ten parameters 62/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 2 mutating operations with no state check
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 85Steps. 96 steps, 1 vague phrases
- 100Failures and branches. 24 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3287 tokens
- 100Progress reporting. Reports progress
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)
- +1No license
- +2Single-language instructions
- +3Description length 682: enough signal without eating the budget
- +4Structure: 25 headings
- +3Step-by-step instructions: 96 items
- +3Output format is stated explicitly
- +4Has examples (9 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 81.