CC sber-business
Работа со СберБизнес API: пошаговый онбординг со ссылками, Zero-Knowledge Vault (AES-256-GCM), выбор канала уведомлений (Telegram/MAX/Discord/Slack/Webhook), остатки, платежки, ссылка на подпись, выгрузка PDF со штампом банка.
Работа со СберБизнес API: пошаговый онбординг со ссылками, Zero-Knowledge Vault (AES-256-GCM), выбор канала уведомлений (Telegram/MAX/Discord/Slack/Webhook)…
As a process C 57/100 · Has gaps — weak spots: result and completion, inputs and preconditions, failures and branches
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 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.
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
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- 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
✓ No critical or high findings
Medium and low: 7
-
medium Obfuscation
uni-mixed-script-wordreferences/HUMAN_GUIDE.md:9Word mixing Latin and Cyrillic letters (homoglyph obfuscation) (1 occurrence)1. Откройте интернет-банк [СберБизнес (sbi.sberbank.ru)](https://sbi.sberbank.ru) и авторизуйтесь по СберID, логину/паролю или токену.
-
medium Exfiltration
exfil-webhook-urlSKILL.md:13Webhook / callback URL commonly used for exfiltration (verify the destination) (the skill's own vendor host)- https://hooks.slack.com (optional Slack webhooks)
vendor-host -
low Exfiltration
exfil-webhook-urlreferences/NOTIFICATION_CHANNELS.md:57Webhook / callback URL commonly used for exfiltration (verify the destination) (placeholder value)"webhookUrl": "https://discord.com/api/webhooks/..."
placeholder -
low Exfiltration
exfil-webhook-urlreferences/NOTIFICATION_CHANNELS.md:74Webhook / callback URL commonly used for exfiltration (verify the destination) (placeholder value)"webhookUrl": "https://hooks.slack.com/services/..."
placeholder -
low Exfiltration
exfil-webhook-urlscripts/config.example.json:22Webhook / callback URL commonly used for exfiltration (verify the destination) (test fixture / example file; the skill's own vendor host; quoted — discussed, not commanded)"webhookUrl": "https://discord.com/api/webhooks/YOUR/WEBHOOK/URL"
fixturevendor-hostquoted -
low Exfiltration
exfil-webhook-urlscripts/config.example.json:26Webhook / callback URL commonly used for exfiltration (verify the destination) (test fixture / example file; the skill's own vendor host; quoted — discussed, not commanded)"webhookUrl": "https://hooks.slack.com/services/YOUR/WEBHOOK/URL"
fixturevendor-hostquoted -
low Exfiltration
exfil-webhook-urlscripts/poller.py:119Webhook / callback URL commonly used for exfiltration (verify the destination) (the skill's own vendor host; quoted — discussed, not commanded)url = f"https://api.telegram.org/bot{token}/sendMessage"vendor-hostquoted
Files scanned: 11. 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")
Process rating: all ten parameters 57/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 3 mutating operations with no state check
- 70When it triggers. States when to use, but not when not to
- 100Tools and files. Tools declared in frontmatter
- 100Steps. 29 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1937 tokens
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
- -31 of 5 scripts are never mentioned in SKILL.md
- +1No license
- +3Description length 226: enough signal without eating the budget
- +4Structure: 12 headings
- +3Step-by-step instructions: 29 items
- +4Has examples (6 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
- +2Bilingual instructions (RU + EN)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.