SKILLEMALL.ai
GitHub ↗ RU Sign in

AC sentinel-transaction-safety

Pre-execution transaction safety check for autonomous AI agents — one capability of the SENTINEL trust oracle (sentinel-agent.dev). Before signing an on-chain transaction on Base, an agent calls SENTINEL and receives a SAFE / UNSAFE / UNKNOWN verdict, a SENTINEL Score (0-100, grade AAA-D), and an ed25519-signed receipt. Checks contract security (GoPlus token security) and execution simulation (Alchemy eth_call). Pay-per-call via x402 on Base — pricing is tiered by transaction value (from $0.005 USDC for small txs, scaling up for higher-value ones). No accounts, no API keys, no SDK required. No free trial. SENTINEL also exposes a separate counterparty-trust capability (behavioral 0-100 scoring); see "SENTINEL is one oracle with two capabilities" below.

ClawHub Agent Skills author: teodorofodocrispin-cmyk v1.1.0 MIT-0 2 files body ≈ 2 860 tokens Open the sourceclawhub.ai↗ analyzed 2 d ago

Pre-execution transaction safety check for autonomous AI agents — one capability of the SENTINEL trust oracle (sentinel-agent.dev). Before signing an on-chain…

As a process C 51/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions

IntegrationGitHubSupabaseAI and agentstype and topics are labelled automatically from the skill text
Runs in: OpenClaw Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) Goose OpenCode
JSON
Technical rating
A
91/100
safety, quality, tests
Safety 60%
95
Quality 40%
85
Run on models
none yet
Process rating
C
51/100
Has gaps
Result and completion w 14
0
Inputs and preconditions w 11
0
Progress reporting w 2
0
the three weakest of ten parameters · all ten

How to improve

    For the model run — optional
    • 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 · 5

    ✓ No critical or high findings

    Medium and low: 5
    • low Secrets in code secret-high-entropy-token SKILL.md:22
      High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
      oracle_onchain: "0x68…813 (Base)"
      quoted
    • low Secrets in code secret-high-entropy-token SKILL.md:27
      High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
      usdc_asset: "0x83…913 (USDC on Base)"
      quoted
    • low Secrets in code secret-high-entropy-token SKILL.md:28
      High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
      pay_to: "0xCf…7E7"
      quoted
    • low Secrets in code secret-high-entropy-token SKILL.md:41
      High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
      > ℹ️ **Name notice:** This is `sentinel-agent.dev` (on-chain oracle `0x68…813`), distinct from unrelated projects sharing the SENTINEL name.
      quoted
    • low Secrets in code secret-high-entropy-token SKILL.md:110
      High-entropy token-like string (may be an id, hash or a credential) (quoted — discussed, not commanded)
      - **On-chain oracle (Base):** `0x68…813`
      quoted

    Files scanned: 1. Evidence is masked. Grey chips explain why severity was lowered.

    Against the Agent Skills spec

    ✓ No remarks against the Agent Skills spec

    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
    • 0Progress reporting. Says nothing while it works
    • 20When it triggers. No condition that starts the skill
    • 30Running it twice. 7 mutating operations with no state check
    • 40Consistency. Frontmatter name (sentinel-transaction-safety) differs from the folder (sentinel-public)
    • 50Failures and branches. 0 branches, has a failure section
    • 100Tools and files. No external tools needed
    • 100Steps. 40 steps
    • 100Execution cost. Instruction body is 2860 tokens
    • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
    • low 12 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)
    • +3Output format is not stated: the model decides each time
    • +2Single-language instructions
    • +3Description length 761: enough signal without eating the budget
    • +4Structure: 18 headings
    • +3Step-by-step instructions: 40 items
    • +4Has examples (4 code blocks)
    • +1License stated

    Quality base 70; lint remarks subtract, signals add up to 100. Result: 85.

    External checks

    ClawHub: clean
    This skill is a disclosed paid transaction-checking integration that sends unsigned transaction details to a remote service before an agent signs a Base transaction.
    LLM: benign (high) · VirusTotal: · 5 Aug 2026