BC browser-secure
Secure browser automation with Chrome profile support, vault integration, approval gates, and comprehensive audit logging. Use for authenticated sites, sensitive operations, or compliance requirements.
As a process C 62/100 · Has gaps — weak spots: result and completion, when it triggers, running it twice
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 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 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
- 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 · 10
✓ No critical or high findings
Medium and low: 10
-
medium Broad scope
meta-broad-allowed-toolsSKILL.md:1Broad tool permissions pre-approved: Bashallowed-tools: Bash
-
medium Exfiltration
intent-browser-credential-storeSKILL.md:217Accesses a browser credential / cookie store (documentation of a security skill)- **Daemon uses a SEPARATE Chrome profile directory** — does not share cookies/logins with your normal Chrome browser. Your normal Chrome tabs are unaffected.
security skill -
low Exfiltration
intent-browser-credential-storedist/browser/secure-session.js:222Accesses a browser credential / cookie store (negated — the text forbids it)// No profile: use Playwright's bundled Chromium (isolated, no cookies from system Chrome)
negated -
low Secrets in code
secret-high-entropy-tokendist/vault/discovery.js:181High-entropy token-like string (may be an id, hash or a credential)const opResult = await inte…ery(domain);
-
low Secrets in code
secret-high-entropy-tokendist/vault/discovery.js:262High-entropy token-like string (may be an id, hash or a credential)async function inte…ery(domain) { -
low Exfiltration
read-dotenvSKILL.md:324Reads a .env file (documentation of a security skill)cp .env.example .env
security skill -
low Exfiltration
read-dotenvSKILL.md:368Reads a .env file (documentation of a security skill)cp .env.example .env
security skill -
low Exfiltration
intent-browser-credential-storesrc/browser/secure-session.ts:288Accesses a browser credential / cookie store (negated — the text forbids it)// No profile: use Playwright's bundled Chromium (isolated, no cookies from system Chrome)
negated -
low Secrets in code
secret-high-entropy-tokensrc/vault/discovery.ts:244High-entropy token-like string (may be an id, hash or a credential)const opResult = await inte…ery(domain);
-
low Secrets in code
secret-high-entropy-tokensrc/vault/discovery.ts:349High-entropy token-like string (may be an id, hash or a credential)async function inte…ery(
Files scanned: 63. 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 62/100
- 0Result and completion. Does not say what the result is
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 15 mutating operations with no state check
- 50Failures and branches. 0 branches, has a failure section
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Execution cost. Instruction body is 4814 tokens
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 53 steps
- 100Consistency. Name and required fields are in place
- low 14 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (9 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)
- +3Output format is not stated: the model decides each time
- -4Absolute local paths (C:\Users, /home/…): not portable
- -2localhost URLs: will not work for another user
- -230 emoji in the instructions: noise for the model
- -31 of 1 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +3Description length 201: enough signal without eating the budget
- +4Structure: 43 headings
- +3Step-by-step instructions: 53 items
- +4Has examples (28 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 73.