DC evm-atlas
Use for targeted EVM chain, account, transaction, RPC, explorer, bridge, and DEX evidence: chain name/ID, native symbol, RouteMesh, wallet balances and DeFi positions via DeBank/Blockscan in Chromium, cross-chain USD portfolio value or net worth, token/NFT holdings/transfers, tx history, funding origin via Etherscan/Blockscout/Chainscout; Across, Bungee, deBridge, Gas.zip, Hop, Layerswap, LayerZero, LI.FI, Relay, Socket, Symbiosis; Uniswap v1-v4, Universal Router, Permit2, 1inch Classic/Fusion/Fusion+, and CoW Swap, CoWSwap, CoW Protocol, or GPv2 swaps, orders, liquidity, approvals, permits, rewards, migrations, wrapping, cancellations, and refunds.
Use for targeted EVM chain, account, transaction, RPC, explorer, bridge, and DEX evidence: chain name/ID, native symbol, RouteMesh, wallet balances and DeFi…
As a process C 50/100 · 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.
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 · 38
✓ No critical or high findings
Medium and low: 38
-
medium Exfiltration
net-credential-usereferences/explorers/blockscout-api.md:136Credential used in a network call (verify the destination is the intended service)curl -s -H "authorization: Bearer $BLOCKSCOUT_API_KEY" \
-
medium Exfiltration
net-credential-usereferences/explorers/blockscout-api.md:160Credential used in a network call (verify the destination is the intended service)curl -s -H "authorization: Bearer $BLOCKSCOUT_API_KEY" \
-
medium Exfiltration
net-credential-usereferences/explorers/blockscout-api.md:164Credential used in a network call (verify the destination is the intended service)curl -s -H "authorization: Bearer $BLOCKSCOUT_API_KEY" \
-
low Secrets in code
secret-high-entropy-tokenreferences/bridges/across.md:101High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)mainnet `SpokePool` is at `0x5c…5C5` (an EIP-1967 proxy; current implementation
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/circle.md:29High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)Ethereum `Toke…ent()` returned `0x6e…Bee` at the checked block
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/circle.md:35High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)- The same transaction shows a matching native USDC mint to `0x6e…Bee`.
detector -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/circle.md:47High-entropy token-like string (may be an id, hash or a credential) (documentation table row)| Ethereum | 1 | `0x6e…Bee` |
table -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/circle.md:48High-entropy token-like string (may be an id, hash or a credential) (documentation table row)| Arbitrum | 42161 | `0x6e…Bee` |
table -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/circle.md:49High-entropy token-like string (may be an id, hash or a credential) (documentation table row)| Avalanche | 43114 | `0x6e…Bee` |
table -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/gaszip.md:24High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)2026-10-01: GasZipV2 at `0x5D…5E0` on Arbitrum Nova (Blockscout-verified source).
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/hop.md:86High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)token. The Arbitrum One `L2_A…per` is at `0x33…417` (verified on-chain, labeled
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/layerswap.md:31High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)Ethereum address `0x2F…efe` is Etherscan-labeled "Layerswap 1", a shared service
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/layerswap.md:135High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)- A transfer touching `0x2F…efe` alone is not proof of attribution to a specific user;
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/relay.md:25High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)- `0xf7…bEF` — Etherscan-labeled "Relay: Solver", Relay's primary solver wallet that
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/relay.md:27High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)- `0xa5…934` — Etherscan-labeled "Reservoir: Relay Receiver", a shared RelayReceiver
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/bridges/relay.md:128High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)- A transfer touching `0xa5…934` or `0xf7…bEF` alone
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/chain-categories.md:54High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)(`0x53…00F`). Post-Curie, the fee on the full signed bytes is
detector -
low Exfiltration
exfil-secret-in-urlreferences/explorers/blockscout-api.md:123Credential passed in a URL query string (normal for some APIs — verify the host is the intended service) (detector / deny-list definition; the skill's own vendor host)curl -s "https://api.blockscout.com/1/api/v2/addresses/0xADDR?apikey=…"
detectorvendor-host -
low Exfiltration
net-credential-usereferences/explorers/blockscout-api.md:123Credential used in a network call (verify the destination is the intended service) (the skill's own vendor host)curl -s "https://api.blockscout.com/1/api/v2/addresses/0xADDR?apikey=…"
vendor-host -
low Exfiltration
net-credential-usereferences/explorers/blockscout-api.md:125Credential used in a network call (verify the destination is the intended service) (the skill's own vendor host)curl -s -H "authorization: Bearer $BLOCKSCOUT_API_KEY" "https://api.blockscout.com/1/api/v2/addresses/0xADDR"
vendor-host -
low Exfiltration
exfil-secret-in-urlreferences/explorers/blockscout-api.md:238Credential passed in a URL query string (normal for some APIs — verify the host is the intended service) (detector / deny-list definition; the skill's own vendor host)curl -s "https://api.blockscout.com/v2/api?chain_id=1&module=…&action=…&address=…&apikey=…"
detectorvendor-host -
low Exfiltration
exfil-secret-in-urlreferences/explorers/blockscout-api.md:268Credential passed in a URL query string (normal for some APIs — verify the host is the intended service) (detector / deny-list definition; the skill's own vendor host)curl -s "https://api.blockscout.com/v2/api?chain_id=1&module=…&action=…&address=…&sort=asc&page=1&offset=10&apikey=…"
detectorvendor-host -
low Exfiltration
exfil-secret-in-urlreferences/explorers/blockscout-api.md:269Credential passed in a URL query string (normal for some APIs — verify the host is the intended service) (detector / deny-list definition; the skill's own vendor host)curl -s "https://api.blockscout.com/v2/api?chain_id=1&module=…&action=…&address=…&sort=asc&page=1&offset=10&apikey=…"
detectorvendor-host -
low Exfiltration
net-credential-usereferences/explorers/etherscan-api.md:67Credential used in a network call (verify the destination is the intended service) (documentation of a security skill; the skill's own vendor host)curl -s "https://api.etherscan.io/v2/api?chainid=1&module=…&action=…&apikey=…"
security skillvendor-host -
low Exfiltration
net-credential-usereferences/explorers/etherscan-api.md:166Credential used in a network call (verify the destination is the intended service) (documentation of a security skill; the skill's own vendor host)curl -s "https://api.etherscan.io/v2/api?chainid=1&module=…&action=…&address=…&tag=…&apikey=…"
security skillvendor-host -
low Exfiltration
net-credential-usereferences/explorers/etherscan-api.md:172Credential used in a network call (verify the destination is the intended service) (documentation of a security skill; the skill's own vendor host)curl -s "https://api.etherscan.io/v2/api?chainid=1&module=…&action=…&address=…&tag=…&apikey=…"
security skillvendor-host -
low Exfiltration
exfil-secret-in-urlreferences/explorers/etherscan-api.md:245Credential passed in a URL query string (normal for some APIs — verify the host is the intended service) (placeholder value)curl -s "https://api.etherscan.io/v2/api?chainid=1&module=…&action=…&address=0x...&page=1&offset=100&apikey=…"
placeholder -
low Exfiltration
net-credential-usereferences/explorers/etherscan-api.md:245Credential used in a network call (verify the destination is the intended service) (documentation of a security skill; the skill's own vendor host)curl -s "https://api.etherscan.io/v2/api?chainid=1&module=…&action=…&address=0x...&page=1&offset=100&apikey=…"
security skillvendor-host -
low Exfiltration
net-credential-usereferences/explorers/etherscan-api.md:285Credential used in a network call (verify the destination is the intended service) (documentation of a security skill; the skill's own vendor host)curl -s "https://api.etherscan.io/v2/api?chainid=1&module=…&action=…&address=…&startblock=0&endblock=…&page=1&offset=100&sort=desc&apikey=$E
security skillvendor-host -
low Secrets in code
secret-high-entropy-tokenreferences/workflows/provider-routing.md:14High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)`0xcA…A11`; do not batch calls whose result depends on the original `msg.sender`.
quoted -
low Secrets in code
secret-high-entropy-tokenscripts/blockscout-detect-plan.sh:26High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)addr="0xde…BAe"
quoted -
low Secrets in code
secret-high-entropy-tokenscripts/check-conformance-fixtures.sh:9High-entropy token-like string (may be an id, hash or a credential) (test fixture / example file; quoted — discussed, not commanded)example_address="0xde…BAe"
fixturequoted -
low Exfiltration
net-credential-usescripts/etherscan-detect-plan.sh:25Credential used in a network call (verify the destination is the intended service) (documentation of a security skill)response=$(curl -fsS "$base?chainid=1&module=getapilimit&action=getapilimit&apikey=$ETHERSCAN_API_KEY")
security skill -
low Exfiltration
net-credential-usescripts/etherscan-detect-plan.sh:70Credential used in a network call (verify the destination is the intended service) (documentation of a security skill)probe=$(curl -fsS "$base?chainid=8453&module=account&action=balance&address=0xde…BAe&tag=latest&apikey=$ETHERSCAN_API_KEY" 2>/dev/null || printf '{"status":"0"}')security skill
A further 4 matches are quotations in this security skill's documentation and are not counted as findings.
Files scanned: 49. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "coordination" - note
frontmatter-keyunknown frontmatter key "skill-dependencies" - note
edit-residuethe text marks something as outdated (lines 61, 75): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 50/100
- 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
- 30Running it twice. 1 mutating operations with no state check
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 100Steps. 26 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2365 tokens
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low The response is described with custom markup (4 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
- +4No input/output examples
- -38 of 9 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +3Description length 657: enough signal without eating the budget
- +4Structure: 4 headings
- +3Step-by-step instructions: 26 items
- +4Reference files are cited in the instructions (1 of 1)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 79.