BB awp
AWP (Agent Working Protocol) — the ONLY source of correct contract addresses, function selectors, API endpoints, and Python scripts for AWP operations. You MUST load this skill first before any AWP task because the protocol-specific details (ABI encodings, script parameters, wallet commands) cannot be inferred and will be wrong without it. Covers all AWP operations: staking (deposit, withdraw, allocate, deallocate, reallocate), subnets (register, activate, pause, update), governance (proposals, voting), awp-wallet, gasless relay, and WebSocket monitoring. Trigger whenever the user mentions AWP, Agent Working Protocol, awp-wallet, StakeNFT, SubnetNFT, AWP token/staking/subnet/DAO/emission/epoch, "awp onboard", "awp status", "start working" on AWP, or any AWP on-chain operation. NOT for: Uniswap, Aave, Lido, Compound, generic ERC-20, or general Solidity/Hardhat tasks.
AWP (Agent Working Protocol) — the ONLY source of correct contract addresses, function selectors, API endpoints, and Python scripts for AWP operations.
As a process B 74/100 · Nearly there — weak spots: 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 skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
How to improve
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- 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 · 6
✓ No critical or high findings
Medium and low: 6
-
medium Dangerous commands
cmd-pipe-to-shellCHANGELOG.md:190Downloads and executes remote code from an unrecognised host (pipe to shell) (quoted — discussed, not commanded)- SKILL.md Step 2: `git clone` → `bash install.sh`（先拉到本地再执行，不用 `curl | bash` 远程管道）
quoted -
medium Dangerous commands
cmd-pipe-to-shellCHANGELOG.md:191Downloads and executes remote code from an unrecognised host (pipe to shell) (quoted — discussed, not commanded)- daemon: all install/update messages use `git clone` + local `install.sh` instead of `curl | bash`
quoted -
medium Dangerous commands
cmd-pipe-to-shellCHANGELOG.md:204Downloads and executes remote code from an unrecognised host (pipe to shell) (quoted — discussed, not commanded)- SKILL.md Step 2: removed hardcoded `curl | bash` install command. Now directs agent to install the AWP Wallet skill (from ClawHub or repo), which handles installation and setup
quoted -
medium Dangerous commands
cmd-pipe-to-shellCHANGELOG.md:211Downloads and executes remote code from an unrecognised host (pipe to shell) (quoted — discussed, not commanded)- SKILL.md Step 2: agent directly runs `curl | bash` to install awp-wallet (not the user)
quoted -
medium Dangerous commands
cmd-pipe-to-shellCHANGELOG.md:219Downloads and executes remote code from an unrecognised host (pipe to shell) (quoted — discussed, not commanded)- **Removed auto-install**: daemon no longer downloads or executes remote install scripts (`curl | bash`). Prints manual install instructions instead
quoted -
low Secrets in code
secret-high-entropy-tokenreferences/commands-subnet.md:85High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)"address": "0xA1…afe",
quoted
Files scanned: 40. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 6360 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 74/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70Execution cost. Instruction body is 6360 tokens
- 100Steps. 32 steps
- 100When it triggers. States when to use and when not to
- 100Failures and branches. 5 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 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 18 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (3 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
- +3Description length 878: 120–800 characters recommended
- -314 of 31 scripts are never mentioned in SKILL.md
- +1No license
- +2Single-language instructions
- +5Description quotes 3 example trigger phrases
- +4Description says when NOT to use the skill
- +4Structure: 39 headings
- +3Step-by-step instructions: 32 items
- +3Output format is stated explicitly
- +4Has examples (58 code blocks)
- +4Reference files are cited in the instructions (3 of 5)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 84.