DC ai-course-agent
Auto-generates AI education courses from natural language requests in Chinese. Detects patterns like "帮我生成6年级数学分数乘除法的课程" and calls Edustem API to create and return a course link. Uses SkillPay for usage-based billing (1 token per course).
As a process C 59/100 · Has gaps — weak spots: when it triggers, inputs and preconditions, failures and branches
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 files contain someone else's key or token. If it is live, your agent will call third-party services under a stranger's identity; if it was revoked, the skill's scripts simply fail. Such a key often arrives with the author's whole workspace, personal data included.
The key is visible to everyone who downloaded the skill and has likely been copied by catalog-scanning bots already. Revoke it now, check bills and access logs, then reissue.
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.
- 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 · 11
-
high Exfiltration
exfil-webhook-urlREADME.md:39Webhook / callback URL commonly used for exfiltration (verify the destination)5. Returns → https://6bb9…app/ai-lesson/{lesson_ref} -
high Exfiltration
exfil-webhook-urlsrc/main-session-integration.md:58Webhook / callback URL commonly used for exfiltration (verify the destination)https://6bb9…app/ai-lesson/3569…659
Medium and low: 9
-
medium Exfiltration
exfil-webhook-urlsrc/edustem-api.ts:4Webhook / callback URL commonly used for exfiltration (verify the destination) (quoted — discussed, not commanded)const API_BASE_URL = "https://6bb9…app/api/v1";
quoted -
medium Exfiltration
exfil-webhook-urlsrc/edustem-api.ts:5Webhook / callback URL commonly used for exfiltration (verify the destination) (quoted — discussed, not commanded)const LESSON_BASE_URL = "https://6bb9…app/ai-lesson";
quoted -
medium Secrets in code
secret-labelled-tokensrc/skillpay.ts:39Labelled token / key literal (vendor format unknown — verify it is not a live credential)const apiKey = 'sk_e…ee5';
-
low Secrets in code
secret-high-entropy-tokenpackage-lock.json:88High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)"integrity": "sha5…cjV+2Z+GK+EEY7…R4x+N3TA…nIr+TMcC…z6Q==",
detector -
low Secrets in code
secret-high-entropy-tokenpackage-lock.json:179High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)"integrity": "sha512-oGB+Uxlg…BKZ+GTy0…lih/NSHS…cSg==",
detector -
low Secrets in code
secret-high-entropy-tokenpackage-lock.json:198High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)"integrity": "sha5…xCA+ORZv…wO5/ywWF…Tag==",
detector -
low Secrets in code
secret-high-entropy-tokenpackage-lock.json:264High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)"integrity": "sha5…JbG+vnnQ…1TK+nxAp…hhm+kzE4…g4g==",
detector -
low Secrets in code
secret-high-entropy-tokenpackage-lock.json:304High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)"integrity": "sha5…MfZ+71RA…HvA==",
detector -
low Secrets in code
secret-password-literalsrc/skillpay.ts:39Hard-coded password / key literal (may be an example)const apiKey = 'sk_e…ee5';
Files scanned: 18. 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 59/100
- 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
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 4 mutating operations with no state check
- 60Result and completion. Output format stated, no completion criterion
- 100Tools and files. No external tools needed
- 100Steps. 9 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 745 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)
- +1No license
- +2Single-language instructions
- +3Description length 238: enough signal without eating the budget
- +4Structure: 12 headings
- +3Step-by-step instructions: 9 items
- +3Output format is stated explicitly
- +4Has examples (5 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.