100/100 (Very Low)

OpenClaw： benign
VirusTotal： benign
StaticScan： clean

OpenClaw： benign

The skill's files, instructions, and requested credential (ABEL_API_KEY) are coherent with a live Abel CAP probing tool: it runs a bundled Python probe against abel.ai, requires a single Abel API key,...

[内容已截断]

VirusTotal： benign VT 报告

静态扫描： clean

No suspicious patterns detected.

README 未提供

{
    "latestVersion": {
        "_creationTime": 1774591304289,
        "_id": "k977vx18xygsk9bc9dzxbcwx2583phmy",
        "changelog": "Changed: Removed `low-signal` wording from the source `causal-abel` planner and report guidance, replacing it with softer bridge-heavy \/ diffuse \/ non-explanatory phrasing so internal routing heuristics do not leak into user-facing language.; Refined direct-graph interpretation guidance so surprising drivers are explained via the security's own attributes before falling back to `weak` or `unresolved` wording.; Trimmed `causal-abel\/agents\/openai.yaml` back toward trigger and routing guidance so detailed execution rules stay in `SKILL.md` and route references.; Tightened the source `causal-abel` prompt so the core guidance is shorter and higher-leverage, with graph-first rules phrased as a small set of primary constraints.; Refined broad ticker-driver guidance so agents anchor to executable tickers, run Abel first, and interpret surprising parents as transmission channels before leaving the graph.; Tightened `causal-abel` prompt priority so direct graph answers now preserve graph facts first instead of replacing them with web-searched narratives.; Reframed Abel graph outputs as high-value PCMCI-style market-data evidence and added guidance for handling surprising drivers as serious transmission signals.; Narrowed direct-graph web grounding so driver lists, parent membership checks, and path facts are usually answered from graph output without forced search.; Updated the report and planner guidance so graph fact, interpretation, and optional web validation stay visibly separate when the graph output is unintuitive.; Reframed `causal-abel` so the default output shape is a compact report rather than a short verdict-only answer.; Updated the direct and proxy routes so executable anchors are observed first, preferring `extensions.abel.observe_predict_resolved_time` before deeper structural traversal.; Changed the planner and probe guidance so non-trivial comparative reads now default to one compact `intervene.do` pressure test after the mechanism is coherent.; Updated the report template so pressure-test coverage is expected by default in longer comparative analyses.",
        "changelogSource": "user",
        "createdAt": 1774591304289,
        "parsed": {
            "clawdis": {
                "homepage": "https:\/\/github.com\/Abel-ai-causality\/Abel-skills",
                "primaryEnv": "ABEL_API_KEY",
                "requires": {
                    "bins": [
                        "python"
                    ],
                    "env": [
                        "ABEL_API_KEY"
                    ]
                }
            }
        },
        "version": "1.0.10"
    },
    "owner": {
        "_creationTime": 0,
        "_id": "s173sq1kpb5aa42yad0z08mavh83h13g",
        "displayName": "ExenVitor",
        "handle": "exenvitor",
        "image": "https:\/\/avatars.githubusercontent.com\/u\/3341682?v=4",
        "kind": "user",
        "linkedUserId": "kn7cv316rbjqxv4404prwm1qqd83f11q"
    },
    "ownerHandle": "exenvitor",
    "skill": {
        "_creationTime": 1774315512173,
        "_id": "kd72axw1qyyqf1ezth06x6bk1583gz6g",
        "badges": [],
        "createdAt": 1774315512173,
        "displayName": "Causal Abel",
        "latestVersionId": "k977vx18xygsk9bc9dzxbcwx2583phmy",
        "ownerPublisherId": "s173sq1kpb5aa42yad0z08mavh83h13g",
        "ownerUserId": "kn7cv316rbjqxv4404prwm1qqd83f11q",
        "slug": "causal-abel",
        "stats": {
            "comments": 0,
            "downloads": 65,
            "installsAllTime": 0,
            "installsCurrent": 0,
            "stars": 0,
            "versions": 3
        },
        "summary": "Use for decision-grade Abel causal reads: explain what is driving a market or company node, how two nodes connect, what changes under intervention, or how a...",
        "tags": {
            "latest": "k977vx18xygsk9bc9dzxbcwx2583phmy"
        },
        "updatedAt": 1774591910081
    }
}