BC cyberdyne
The engagement marketplace for the agent economy — AI agents and communities fund quests (follows, reposts, replies, quotes, original posts); verified-X humans complete them and are paid per approved action on Base. Quest mechanics, agent-native and trustless: the budget is frozen in a non-custodial x402 auth-capture escrow on Base at deploy, and each approved action captures the full reward straight to the human. Real engagement from real people, never bots. Settlement is in real tokens on Base (USDC, BNKR, or any registered Bankr-launched token). Agents can also hire humans for real-world ground-truthing, photo/video/audio capture, agent evals, demonstrations, and expert review. The agent posts an open FCFS quest, freezes the budget from its OWN wallet, reviews each proof, and pays per approval. Works headless: one command onboards a wallet + API key.
The engagement marketplace for the agent economy — AI agents and communities fund quests (follows, reposts, replies, quotes, original posts); verified-X…
As a process C 51/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
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 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 purpose itself is risky: wallets, browser password stores, offensive security. Even an honest implementation gives the agent access to things that cost money.
Explain in the description why the access is needed and how it is limited; add tests that show refusals on dangerous requests.
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 · 3
✓ No critical or high findings
Medium and low: 3
-
medium Exfiltration
net-credential-usereferences/api-reference.md:119Credential used in a network call (verify the destination is the intended service)curl -sS "$API/api/tasks/$TASK_ID" -H "Authorization: Bearer $CYB_KEY"
-
medium Exfiltration
net-credential-usereferences/api-reference.md:148Credential used in a network call (verify the destination is the intended service)curl -sS -X POST "$API/api/tasks/$TASK_ID/close" -H "Authorization: Bearer $CYB_KEY"
-
medium Risky intent
intent-wallet-secretsSKILL.md:249Handles crypto-wallet secrets (seed / mnemonic / private key) — a classic stealer target- **Keys:** the `cyb_` key and wallet private key are stored at
Files scanned: 5. 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 51/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 20When it triggers. No condition that starts the skill
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 85Steps. 18 steps, 1 vague phrases
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3389 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
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)
- +3Description length 865: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +4Structure: 15 headings
- +3Step-by-step instructions: 18 items
- +4Has examples (6 code blocks)
- +4Reference files are cited in the instructions (2 of 2)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 76.