Is Feedoracle Managed Agents safe to install?

SaferSkills independently audited Feedoracle Managed Agents (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

100

35.0 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 0 flagged

Securityscore 100 · 0 findings

✓—All security checks passedNo findings in this category for the latest scan.pass

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

README.md · 1 file

README.md6.8 KB · Markdown

FeedOracle × Claude Managed Agents

Turn compliance questions into signed evidence — with one API call.

FeedOracle provides the compliance evidence layer for Claude's new Managed Agents platform. Connect 5 MCP servers. Get MiCA compliance verdicts, stablecoin risk scores, macro regime signals, and auditable governance — all running autonomously in Anthropic's managed infrastructure.

What you get

MCP Server	Endpoint	Tools	What it does
Compliance	`feedoracle.io/mcp/`	26	MiCA status, evidence packs, peg deviation, reserves, compliance pre-flight
Risk	`feedoracle.io/mcp/risk/sse`	13	7-signal stablecoin risk scoring, custody risk, DEX liquidity
Macro	`feedoracle.io/mcp/macro/sse`	13	Macro regime, Fed watch, yield curve, VIX, recession probability
AgentGuard	`feedoracle.io/guard/mcp/`	17	Audit logging, KYA identity, policy enforcement, secret detection
Bitcoin	`tooloracle.io/btc/mcp/`	10	BTC price, blocks, transactions, mempool, fees, halving

Total: 79 tools available to your agent. No API keys needed.

Quickstart (bash)

export ANTHROPIC_API_KEY="sk-ant-..."
bash quickstart.sh

Quickstart (Python)

pip install anthropic httpx
export ANTHROPIC_API_KEY="sk-ant-..."
python feedoracle_agent.py

How it works

Your App                 Anthropic Cloud              FeedOracle
   │                          │                           │
   ├─ Create Agent ──────────►│                           │
   │  (5 MCP servers)         │                           │
   │                          │                           │
   ├─ Start Session ─────────►│                           │
   │                          ├─ Provision container      │
   │                          │                           │
   ├─ "Is USDC MiCA          │                           │
   │   compliant?" ──────────►│                           │
   │                          ├─ mica_status(USDC) ──────►│
   │                          │◄── {status: compliant} ───┤
   │                          │                           │
   │                          ├─ peg_deviation(USDC) ────►│
   │                          │◄── {deviation: 0.0003} ───┤
   │                          │                           │
   │                          ├─ evidence_bundle(USDC) ──►│
   │                          │◄── {pack_id, signature} ──┤
   │                          │                           │
   │                          ├─ audit_log(decision) ────►│
   │                          │◄── {logged, chain_id} ────┤
   │                          │                           │
   │◄── Signed verdict ───────┤                           │
   │    with evidence pack    │                           │

Agent Definition (curl)

curl https://api.anthropic.com/v1/agents \
  -H "x-api-key: $ANTHROPIC_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "anthropic-beta: managed-agents-2026-04-01" \
  -H "content-type: application/json" \
  -d '{
    "name": "FeedOracle Compliance Agent",
    "model": "claude-sonnet-4-6",
    "system": "You are the FeedOracle Compliance Agent...",
    "tools": [{"type": "agent_toolset_20260401"}],
    "mcp_servers": [
      {"type": "url", "url": "https://feedoracle.io/mcp/", "name": "feedoracle-compliance"},
      {"type": "url", "url": "https://feedoracle.io/mcp/risk/sse", "name": "feedoracle-risk"},
      {"type": "url", "url": "https://feedoracle.io/mcp/macro/sse", "name": "feedoracle-macro"},
      {"type": "url", "url": "https://feedoracle.io/guard/mcp/", "name": "feedoracle-agentguard"},
      {"type": "url", "url": "https://tooloracle.io/btc/mcp/", "name": "tooloracle-btc"}
    ]
  }'

Example queries

MiCA Compliance Check:

"Run a full MiCA compliance pre-flight for USDC. Generate a signed evidence pack."

Macro Regime Detection:

"What is the current macro regime? Should my trading agent be risk-on or risk-off?"

Stablecoin Portfolio Scan:

"Compare risk scores of USDC, USDT, DAI, RLUSD. Rank by MiCA suitability for an EU fund."

Agent Governance:

"Register this agent via KYA, then audit-log a compliance decision for USDT with full evidence chain."

Adding more MCP servers

FeedOracle/ToolOracle operates 85+ MCP servers with 860+ tools. Add any of these to your agent:

{"type": "url", "url": "https://feedoracle.io/mcp/macro/sse", "name": "macro"}
{"type": "url", "url": "https://tooloracle.io/sol/mcp/", "name": "solana"}
{"type": "url", "url": "https://tooloracle.io/eth/mcp/", "name": "ethereum"}
{"type": "url", "url": "https://tooloracle.io/btc/mcp/", "name": "bitcoin"}
{"type": "url", "url": "https://tooloracle.io/arb/mcp/", "name": "arbitrum"}
{"type": "url", "url": "https://tooloracle.io/base/mcp/", "name": "base"}
{"type": "url", "url": "https://tooloracle.io/sui/mcp/", "name": "sui"}
{"type": "url", "url": "https://tooloracle.io/ton/mcp/", "name": "ton"}
{"type": "url", "url": "https://tooloracle.io/hedera/mcp/", "name": "hedera"}
{"type": "url", "url": "https://feedoracle.io/flight/mcp/", "name": "flights"}
{"type": "url", "url": "https://feedoracle.io/hotel/mcp/", "name": "hotels"}
{"type": "url", "url": "https://feedoracle.io/jobs/mcp/", "name": "jobs"}
{"type": "url", "url": "https://feedoracle.io/shop/mcp/", "name": "shopping"}
{"type": "url", "url": "https://feedoracle.io/news/mcp/", "name": "news"}

Full catalog: feedoracle.io | tooloracle.io

Cost

Component	Price
Managed Agents runtime	$0.08 / session hour
Claude Sonnet 4.6 tokens	Standard API pricing
FeedOracle MCP servers	Free (no API key needed)

Architecture

Agent → Reusable config (model + prompt + tools + MCP servers)
Environment → Container with network access to FeedOracle endpoints
Session → Running instance, maintains state across tool calls
Events → SSE stream of agent actions, tool calls, and results

License

MIT — Built by FeedOracle Technologies

Feedoracle Managed Agents— agent skill

Is Feedoracle Managed Agents safe to install?

FeedOracle × Claude Managed Agents

What you get

Quickstart (bash)

Quickstart (Python)

How it works

Agent Definition (curl)

Example queries

Adding more MCP servers

Cost

Architecture

Links

License

Audit the pieces. Scan the whole. Decide.