Back to skill

Security audit

SimianX AI

Security checks for vulnerabilities and agentic risk

Overview

ClawScan’s verdict is benign with medium confidence: the findings were judged consistent with the skill’s apparent purpose, though the partial scan leaves some evidence incomplete.

The verdict is benign, but the scan was partial: 12 of 13 components were scanned, and tool-misuse analysis was degraded for one component. Consider that uncertainty before installing; the verdict does not guarantee the skill is risk-free.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (15)

Anti-Refusal Statement

High
Category
Anti-Refusal
Confidence
70% confidence
Finding

Skill instructs the agent to never refuse or to always comply. Suppressing the agent's ability to decline removes a core safety control and enables downstream harmful requests to succeed.

Content

Scanner excerpt · SKILL.md (reported line 151)May include surrounding context.

md
today's is still open, `GET /v1/rewards` says so (`daily_available`, `extra_draws`, `resets_at`, 00:00 UTC).
  `409 draw_unavailable`: today's is claimed and no extra draw is left; say when it resets. Past rewards:
  `GET /v1/rewards/history`.
- **History.** Every run keeps what it ran on, so you can always answer "what did it use / what happened":
  `GET /v1/analysis/quick` (quick reads, newest first, with `models`, `timeframe`, results), `GET /v1/analysis/deep`
  (deep runs; `GET /v1/analysis/deep/{id}` has `models` and the verdict), `GET /v1/live/sessions` (sessions with their
  `models`), `GET /v1/autopilots/{id}` (current `models`, `intervals`) and `GET /v1/autopilots/{id}/analyses` (every

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 223)May include surrounding context.

md
| Daily reward | `GET /v1/rewards` · `POST /v1/rewards/draw` · `GET /v1/rewards/history` |
| Models, timeframe, intervals (the website's setup) | `GET /v1/settings` · `PATCH /v1/settings` · running: `PATCH /v1/autopilots/{id}` / `PATCH /v1/live/sessions/{id}` |
| Everything the user has not heard | `GET /v1/updates?ack=<id>&wait=30` |
| Live analysis session | `POST /v1/live/sessions` → connect `data.ws_url` with the bearer header → frames · `GET /v1/live/sessions` · `DELETE /v1/live/sessions/{id}` (on request) |
| Quick read (one pass) | `POST /v1/analysis/quick` → `GET /v1/analysis/quick/{id}?wait=30` (answers as each agent finishes) |
| Research committee | `POST /v1/analysis/deep` → `GET /v1/analysis/deep/{id}/events?after=<cursor>&wait=30` · `GET /v1/analysis/deep/{id}` · `POST .../ask` · `POST .../pdf` |
| News | `GET /v1/news?symbols=BTC,NVDA&since=<iso>` · `GET /v1/news/{id}` · `GET /v1/news/stream` (SSE, Plus+) |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · references/live.md (reported line 38)May include surrounding context.

md
- `PATCH /v1/live/sessions/{id}` `{"models": {"decision": "<id>"}}` switches models while it runs (paid plans; the next
  analysis uses them; the saved setup in `GET /v1/settings` stays as it is).
- `GET /v1/live/sessions` lists the key's sessions (`?status=running`), `GET /v1/live/sessions/{id}` one of them,
  `DELETE /v1/live/sessions/{id}` (on request) stops it (also stops the analysis; the socket receives `session_end`).
  In `GET`, `PATCH` and `DELETE /v1/live/sessions/{id}`, `{id}` is the live session's `id` or the `session_id` in its
  `ws_url`. An autopilot's `session_id` is not a live session: there the answer is 404 with `data.autopilot_id` and
  `data.url`; read it with `GET /v1/autopilots/{id}`.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 227)May include surrounding context.

md
| Quick read (one pass) | `POST /v1/analysis/quick` → `GET /v1/analysis/quick/{id}?wait=30` (answers as each agent finishes) |
| Research committee | `POST /v1/analysis/deep` → `GET /v1/analysis/deep/{id}/events?after=<cursor>&wait=30` · `GET /v1/analysis/deep/{id}` · `POST .../ask` · `POST .../pdf` |
| News | `GET /v1/news?symbols=BTC,NVDA&since=<iso>` · `GET /v1/news/{id}` · `GET /v1/news/stream` (SSE, Plus+) |
| News watch (server side, offline too) | `POST /v1/news/watches {"symbols", "min_value"}` · `GET /v1/news/watches` · `DELETE /v1/news/watches/{id}` (on request) |
| Market Pulse | `GET /v1/pulse/events` · `GET /v1/pulse/summary` |
| Search before starting | `GET /v1/instruments/search?q=` (best match first; use `symbol`); several names at once: `?symbols=NVDA,SOXL,PEPE` (up to 20, one call) · `GET /v1/market/status` |
| Crypto data | `GET /v1/quotes?symbols=` · `GET /v1/market/{symbol}/candles` · `/indicators` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 235)May include surrounding context.

md
| Start / list autopilots | `POST /v1/autopilots` · `GET /v1/autopilots` |
| Follow an autopilot | the updates feed, or `GET /v1/autopilots/{id}/events?after=<cursor>&wait=30` (long-poll, free) or `&stream=1` (SSE) |
| Autopilot history | `GET /v1/autopilots/{id}/analyses` · `/positions` · `/logs` |
| Stop / restart / archive | `POST /v1/autopilots/{id}/stop` · `/start` · `DELETE /v1/autopilots/{id}` (on request) |
| Trade the paper book by hand | `POST /v1/autopilots/{id}/actions` `{"action": "open|add|reduce|close|set_levels|place_order|replace_order|cancel_order|set_timeframe", …}` (server prices) |
| Run one agent now | `POST /v1/autopilots/{id}/run {"agent": "decision"}` → result in the feed and the log |
| Webhooks (code clients) | `POST /v1/webhooks {"url", "events", "secret"}` · `GET` · `DELETE /v1/webhooks/{id}` (on request) · `POST /v1/webhooks/{id}/test` · `GET /v1/webhooks/{id}/failures` |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · references/autopilots.md (reported line 195)May include surrounding context.

md
- `PATCH /v1/autopilots/{id}` with any of `timeframe`, `models`, `intervals`, `timing_gate`, `language`, `ai_position_control`.
  `timeframe` behaves like the `set_timeframe` action (closes the book first); `ai_position_control: false` cancels the
  resting order and closes the open position (`close_reason: user_disabled`, no notification, still in the event log).
- `POST /v1/autopilots/{id}/stop`, `POST /v1/autopilots/{id}/start` (restart, same checks as create), `DELETE /v1/autopilots/{id}` (on request) (archive; must be stopped)

Stops from the web app also appear as `autopilot.stopped` in the log, so a runner always learns about them.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 238)May include surrounding context.

md
| Stop / restart / archive | `POST /v1/autopilots/{id}/stop` · `/start` · `DELETE /v1/autopilots/{id}` (on request) |
| Trade the paper book by hand | `POST /v1/autopilots/{id}/actions` `{"action": "open|add|reduce|close|set_levels|place_order|replace_order|cancel_order|set_timeframe", …}` (server prices) |
| Run one agent now | `POST /v1/autopilots/{id}/run {"agent": "decision"}` → result in the feed and the log |
| Webhooks (code clients) | `POST /v1/webhooks {"url", "events", "secret"}` · `GET` · `DELETE /v1/webhooks/{id}` (on request) · `POST /v1/webhooks/{id}/test` · `GET /v1/webhooks/{id}/failures` |
| Models you may set per agent | `GET /v1/models` (public) |
| Tell us something | `POST /v1/feedback` (read, not answered) |

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · references/billing.md (reported line 63)May include surrounding context.

md
uses it only on the user's request. A key may create keys (`POST /v1/keys`) only within its own scopes. Agents
  never receive `billing:manage`.
- `GET /v1/keys`, `GET /v1/keys/{id}` (with 30-day usage), `PATCH /v1/keys/{id}` (name, budgets, scopes, expires_at,
  ip_whitelist), `DELETE /v1/keys/{id}` (revoke, on request). The token is returned once at creation.
- `GET /v1/usage?days=7` → requests and points by key, endpoint and day.
- Send `Idempotency-Key` on POST / PATCH / DELETE to make retries safe (24 h replay, `X-Idempotent-Replay: true`,
  never charged again; a replayed key creation never repeats the token, a webhook its secret). A retry that arrives

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · references/developers.md (reported line 47)May include surrounding context.

"acks": true, "this_key_only": false} → 201 {"webhook": {"id", "url", "events", "active", "acks", …}, "secret": "", "signature_header": "X-SimianX-Signature: t=,v1=<hex hmac_sha256(secret, '.')>"} GET /v1/webhooks · GET /v1/webhooks/{id} · PATCH /v1/webhooks/{id} {url?, events?, acks?, active?} · DELETE /v1/webhooks/{id} (on request) POST /v1/webhooks/{id}/test → one signed "test" event to your URL now GET /v1/webhooks/{id}/failures → the dead letters: update id, attempt, response status, error

text

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · references/live.md (reported line 39)May include surrounding context.

md
analysis uses them; the saved setup in `GET /v1/settings` stays as it is).
- `GET /v1/live/sessions` lists the key's sessions (`?status=running`), `GET /v1/live/sessions/{id}` one of them,
  `DELETE /v1/live/sessions/{id}` (on request) stops it (also stops the analysis; the socket receives `session_end`).
  In `GET`, `PATCH` and `DELETE /v1/live/sessions/{id}`, `{id}` is the live session's `id` or the `session_id` in its
  `ws_url`. An autopilot's `session_id` is not a live session: there the answer is 404 with `data.autopilot_id` and
  `data.url`; read it with `GET /v1/autopilots/{id}`.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · references/news.md (reported line 40)May include surrounding context.

"include_pulse": false, "quiet_hours": {"start": "23:00", "end": "07:00", "tz": "America/Los_Angeles"}} → 201 {"watch": {"id": "…", "symbols": [...], "min_value": "medium", "active": true, …}, "next_step": "…"} GET /v1/news/watches → the key's watches (symbols, when the last match was, how many) DELETE /v1/news/watches/{id} → stop (when the user asks)

text

`language` is the language of the user who asked: `"zh-CN"` for "帮我盯 NVDA 的新闻", `"en"` for the English request.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
80% confidence
Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · references/sectors.md (reported line 113)May include surrounding context.

md
- `POST /v1/sectors/{sector}/follows` `{"type": "stock", "ticker": "NVDA"}` · `{"type": "asset", "slug": "<slug>"}`
  · `{"type": "calendar", "id": "<calendar item id>", "reminder_offsets": [7, 1, 0]}` (days before; omit to inherit
  the subscription's offsets). Idempotent per target: the same target twice returns the existing follow.
- `DELETE /v1/sectors/{sector}/follows/{id}` (on request) → removes it. The subscription stays; it is managed on the site.

## Poll loop

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 256)May include surrounding context.

bash
# 1. a quick read: the four live agents on BTC 15m, once
curl -s $API/v1/analysis/quick -H "Authorization: Bearer $SIMIANX_API_KEY" -H 'Content-Type: application/json' \
  -d '{"asset":"crypto","symbol":"BTC","timeframe":"15m"}'
# → 202, data.run.id, data.next_step ("keep following…")

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/autopilots.md (reported line 128)May include surrounding context.

book is yours alone (the agent still analyses and fires signal.fired). Paper only, and only when the user asks:

bash
curl -s $API/v1/autopilots/$ID/actions -H "Authorization: Bearer $SIMIANX_API_KEY" -H 'Content-Type: application/json' \
  -d '{"action":"open","side":"long","notional_usd":1000,"stop_loss":96000,"take_profit":108000}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
70% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · references/developers.md (reported line 119)May include surrounding context.

md
API = "https://service.simianx.ai"; KEY = os.environ["SIMIANX_API_KEY"]
H = {"Authorization": f"Bearer {KEY}", "X-SimianX-Skill": "1.0.16", "Content-Type": "application/json"}

s = requests.post(f"{API}/v1/live/sessions", headers={**H, "Idempotency-Key": "live-btc-15m-1"},
                  json={"asset": "crypto", "symbol": "BTC", "timeframe": "15m"}).json()["data"]

async def follow():