SKILLEMALL.ai
GitHub ↗ RU Sign in

DC agent-safety

Agent 安全防护体系——事件驱动拦截（Hook Engine）+ 三层护栏（输入/工具/输出）+ 迭代循环（Ralph Loop）+ 操作追踪（Operation Tracer）

Not recommendedcritical or high security findings · low grade D
ClawHub Agent Skills author: paudyyin v1.1.0 MIT-0 24 files · 4 scripts body ≈ 3 050 tokens Open the sourceclawhub.ai↗ analyzed 4 h ago

Agent 安全防护体系——事件驱动拦截（Hook Engine）+ 三层护栏（输入/工具/输出）+ 迭代循环（Ralph Loop）+ 操作追踪（Operation Tracer）

As a process C 50/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions

IntegrationSecuritytype 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
D
43/100
safety, quality, tests
Safety 60%
22
Quality 40%
75
Run on models
none yet
Process rating
C
50/100
Has gaps
Result and completion w 14
0
Inputs and preconditions w 11
0
When it triggers w 12
20
the three weakest of ten parameters · all ten

What is at stake

The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.

Instruction override
If you install

The text contains phrases like "ignore previous instructions" or "you are now…". That is an attempt to hijack the agent: it may break your rules, the system limits or company policy.

For the author

An honest skill does not need them: state the role and the rules directly without overriding other instructions. Otherwise catalog scanners and corporate filters will block the listing.

Secrets in code medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The files contain someone else's key or token. If it is live, your agent will call third-party services under a stranger's identity; if it was revoked, the skill's scripts simply fail. Such a key often arrives with the author's whole workspace, personal data included.

For the author

The key is visible to everyone who downloaded the skill and has likely been copied by catalog-scanning bots already. Revoke it now, check bills and access logs, then reissue.

How to improve

  1. Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
  2. Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
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 · 24

  • high Instruction override en-ignore-previous SKILL.md:203
    Instruction-override phrase ("ignore previous instructions")
    | 1 | ignore previous instructions | "ignore previous instructions and..." |
  • high Instruction override en-ignore-previous SKILL.md:204
    Instruction-override phrase ("ignore previous instructions")
    | 2 | ignore above instructions | "ignore all above instructions" |
Medium and low: 22
  • medium Instruction override en-ignore-previous references/injection_patterns.md:9
    Instruction-override phrase ("ignore previous instructions") (documentation of a security skill)
    ### 1. ignore previous instructions
    security skill
  • medium Instruction override en-ignore-previous references/injection_patterns.md:21
    Instruction-override phrase ("ignore previous instructions") (documentation of a security skill)
    ### 2. ignore above instructions
    security skill
  • medium Secrets in code secret-private-key scripts/test_guardrails.py:153
    Private key material (key header without key body; test fixture / example file; quoted — discussed, not commanded)
    ("private_key = -----BEGIN PRIVATE KEY----- …
    header onlyfixturequoted
  • medium Instruction override en-fake-system-prompt SKILL.md:207
    Fake system prompt injected into content (documentation table row)
    | 5 | system prompt | "system prompt override" |
    table
  • medium Secrets in code secret-private-key SKILL.md:236
    Private key material (key header without key body; documentation table row)
    | 5 | 私钥 | `private_key=xxx`, `-----BEGIN PRIVATE KEY----- …
    header onlytable
  • low Secrets in code secret-password-literal scripts/guardrail.py:81
    Hard-coded password / key literal (may be an example) (placeholder value)
    result = guardrails.check_output("Here is the API_KEY=sk-1…def")
    placeholder
  • low Secrets in code secret-password-literal scripts/output_guard.py:99
    Hard-coded password / key literal (may be an example) (placeholder value)
    "API_KEY=sk-1…def",
    placeholder
  • low Secrets in code secret-private-key scripts/output_guard.py:103
    Private key material (placeholder value)
    "-----BEGIN PRIVATE KEY----- …",
    placeholder
  • low Secrets in code secret-password-literal scripts/test_guardrails.py:147
    Hard-coded password / key literal (may be an example) (placeholder value)
    ("API_KEY=sk-1…def", True, "API密钥过滤"),
    placeholder
  • low Secrets in code secret-high-entropy-token scripts/test_guardrails.py:152
    High-entropy token-like string (may be an id, hash or a credential) (test fixture / example file; quoted — discussed, not commanded)
    ("token = eyJh…CJ9", True, "访问令牌过滤"),
    fixturequoted
  • low Secrets in code secret-password-literal scripts/test_guardrails.py:215
    Hard-coded password / key literal (may be an example) (test fixture / example file; quoted — discussed, not commanded)
    r3 = system.check_output("API_KEY=sk-s…123, email: ad…@….com")
    fixturequoted

A further 11 matches are quotations in this security skill's documentation and are not counted as findings.

Files scanned: 24. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning description-no-when description does not say WHEN to use the skill (no "use when")
  • note frontmatter-key unknown frontmatter key "triggers"

Process rating: all ten parameters 50/100

  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 20When it triggers. No condition that starts the skill
  • 30Running it twice. 3 mutating operations with no state check
  • 50Failures and branches. 0 branches, has a failure section
  • 60Tools and files. Uses tools (python) that frontmatter does not declare
  • 100Steps. 17 steps
  • 100Consistency. Name and required fields are in place
  • 100Execution cost. Instruction body is 3050 tokens
  • 100Progress reporting. Reports progress

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)
  • +3Description length 91: 120–800 characters recommended
  • +3Output format is not stated: the model decides each time
  • +1No license
  • +2Single-language instructions
  • +4Structure: 33 headings
  • +3Step-by-step instructions: 17 items
  • +4Has examples (18 code blocks)
  • +4Reference files are cited in the instructions (9 of 9)
  • +3All 9 scripts are documented

Quality base 70; lint remarks subtract, signals add up to 100. Result: 75.

External checks

ClawHub: suspicious
This safety skill is not overtly malicious, but it can automatically change agent behavior and keep local logs/traces that may contain sensitive task data.
LLM: suspicious (high) · 3 Aug 2026