Back to skill

Security audit

OpenClaw Launch

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill benign with medium confidence: its reported external service requests appear aligned with its purpose, though the scan was partial and noted a missing local reference.

The verdict is benign, not risk-free, and has medium confidence. Skillspector coverage is partial: a local path-like reference could not be resolved, so the evidence is incomplete. Consider whether you are comfortable with the reported requests to the external service before installing.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (12)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% 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 24)May include surrounding context.

md
ir own shell or agent config (for example `export OPENCLAW_LAUNCH_API_KEY=...`). Do not ask them to paste the key into the chat, and do not write it into any file yourself.
- Recommend a dedicated key even if the user already has one. A key can only see and control the bots that same key created, so an agent-only key cannot reach bots made by the user's storefront or other keys, and revoking it never disrupts them. It does share the account's balance, plan limits and model funding (see "What a key can cost").
- Never print, echo, log or summarize the key, never write it to a file, a commit, an issue or a bot message, never send it to any host other than `openclawlaunch.com`, and never run these commands with `set -x`, `curl -v` or `--trace`.

Every call sends the key the same way: `printf` writes the header into curl's config on stdin (`-K -`), so the key never appears in your command text or in curl's process arguments. `-q` comes first so a local `.curlrc` cannot change the request.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% 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 38)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sS -K - "https://openclawlaunch.com/api/v1/instances"

That is also the read-only check that the key works. 401 unauthorized means the key is missing, revoked or mistyped. 403 wrong_key_surface means it is a coding-workspace key; the user needs a key made for Build & Resell bots.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% 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 66)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sS -K - -X POST "https://openclawlaunch.com/api/v1/instances" \
      -H "Content-Type: application/json" \
      -d '{"framework":"openclaw","size":"lite","assistant_name":"Support Bot","client_ref":"skill-<short-unique-id>"}'

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% 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 89)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sS -K - "https://openclawlaunch.com/api/v1/instances/$BOT_ID"

Returns {"data": {"id", "status", "framework", "size", "assistant_name", ...}}. A new bot goes from starting to running, usually in seconds and at most a couple of minutes. Poll every 10 seconds, at most 20 times, then report the last status instead of waiting forever. GET /instances lists every bot this key owns as {"data": [...]}.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% 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 139)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sS -K - -X POST "https://openclawlaunch.com/api/v1/instances/$BOT_ID/stop"

Use /start or /restart the same way. These are reversible. Remember that a stopped bot is still billed on Scale.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% 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 150)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sS -K - -X DELETE "https://openclawlaunch.com/api/v1/instances/$BOT_ID"

Returns {"data": {"id": "...", "deleted": true}}.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% 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 98)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sSN -K - -X POST "https://openclawlaunch.com/api/v1/responses" \
      -H "Content-Type: application/json" \
      -d '{"instance_id":"<BOT_ID>","input":"Hello, what can you do?","session_id":"skill-demo"}'

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% 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 120)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\ndata = "{\\"action\\":\\"connect\\",\\"platform\\":\\"telegram\\",\\"bot_token\\":\\"%s\\"}"\n' \
    "$OPENCLAW_LAUNCH_API_KEY" "$TELEGRAM_BOT_TOKEN" \
  | curl -q -sS -K - -X POST "https://openclawlaunch.com/api/v1/instances/$BOT_ID/channels" \
      -H "Content-Type: application/json"

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
60% 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 128)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sS -K - -X POST "https://openclawlaunch.com/api/v1/instances/$BOT_ID/channels" \
      -H "Content-Type: application/json" \
      -d '{"action":"pairing_approve","platform":"telegram","code":"<CODE>"}'

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 66)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sS -K - -X POST "https://openclawlaunch.com/api/v1/instances" \
      -H "Content-Type: application/json" \
      -d '{"framework":"openclaw","size":"lite","assistant_name":"Support Bot","client_ref":"skill-<short-unique-id>"}'

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 98)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sSN -K - -X POST "https://openclawlaunch.com/api/v1/responses" \
      -H "Content-Type: application/json" \
      -d '{"instance_id":"<BOT_ID>","input":"Hello, what can you do?","session_id":"skill-demo"}'

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 128)May include surrounding context.

bash
printf 'header = "Authorization: Bearer %s"\n' "$OPENCLAW_LAUNCH_API_KEY" \
  | curl -q -sS -K - -X POST "https://openclawlaunch.com/api/v1/instances/$BOT_ID/channels" \
      -H "Content-Type: application/json" \
      -d '{"action":"pairing_approve","platform":"telegram","code":"<CODE>"}'