Audit Report: skill-vetter — 🟠 D (34/100)

Audited by TAR Engine · 2026-07-30 · Report format v0.2

Reading note: this edition uses gpt-4o-mini as the victim model and the same model as the adversarial-fuzz judge. Findings reflect missing defenses in the SKILL.md itself — not a verdict on any specific victim model. The remediation belongs in SKILL.md, not in the model.

Source: https://github.com/dvcrn/openclaw-skills-marketplace/blob/main/plugins/spclaudehome--skill-vetter/skills/skill-vetter/SKILL.md

Verdict: High risk — 6 high-severity issues need author attention before deploying to a shared environment.

What this skill does

Auditor's read (LLM-generated): The skill-vetter conducts a security assessment of AI agent skills before installation by checking the source, reviewing the code for red flags, and evaluating permission scopes. It generates a detailed vetting report that includes metrics, identified risks, and a verdict on whether the skill is safe to install. The skill utilizes commands to fetch repository statistics and file lists from GitHub for thorough evaluation.

Author description: Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.

Observed: skill-vetter is 6 top-level sections (When to Use, Vetting Protocol, Output Format, Quick Vet Commands, Trust Hierarchy, …); ~132 lines of instructions, makes outbound network calls, concise body.

Frontmatter facts:

• Body size: 132 lines / 3677 chars

Score breakdown by category

Each category gets its own sub-score. A category with no rule hits gets 100; a category with a single critical finding drops to 80.

Historical baseline (same-skill comparison)

• Prior audits on record: 2 (first 2026-07-30T20:26:37.773965Z, most recent prior 2026-07-30T20:26:54.040037Z)
• Score statistics: mean 34.0 ± 0.0 (range 34–34) (normal band: 31.0 – 37.0)
• This audit vs last: 0 (➡️ stable)
• Top recurring findings across history:
• SEM-006 — hit in 2 of 2 prior audits (100.0%)
• SEM-008 — hit in 2 of 2 prior audits (100.0%)
• SEM-002 — hit in 2 of 2 prior audits (100.0%)
• SEM-003 — hit in 2 of 2 prior audits (100.0%)
• SEM-004 — hit in 2 of 2 prior audits (100.0%)

Baseline assumes the skill's name + description haven't changed. A rename or rewrite starts a fresh baseline.

Findings

8 rules matched. Each finding below cites the matched line and a remediation hint.

1. 🟠 SEM-006 — credential_handling_unsafe (HIGH)

• Category: Credential exposure
• Why this matched: The skill's request for sensitive credentials without secure handling could lead to exposure or misuse of user data.
• Rule intent: Credentials/tokens handled in ways that risk exposure (logging, echoing, including in prompts)
• Matches in document: 1

Evidence (1 of 1 match):

Line 39:

• Requests credentials/tokens/API keys

Suggested fix: Ensure that any requests for credentials are handled securely, using encryption and secure storage practices, and provide clear guidelines on how to manage sensitive information.

2. 🟠 SEM-008 — external_payload_blind_trust (HIGH)

• Category: Malicious payload signatures
• Why this matched: The skill fetches data from external GitHub APIs without validating the content, which could lead to executing malicious code if the API is compromised.
• Rule intent: Trusts external content (downloaded file, remote prompt template, third-party output) without validation
• Matches in document: 1

Evidence (1 of 1 match):

Line 111:

# Check repo stats

Suggested fix: Implement validation checks on the data fetched from external sources to ensure its integrity and safety before processing it.

3. 🟠 SEM-002 — ambiguous_instruction (HIGH)

• Category: Prompt injection / scope override
• Why this matched: The instruction to vet skills from ClawdHub is vague and could lead a user to install malicious skills if they misinterpret the vetting process.
• Rule intent: Ambiguous instruction that could be exploited as a prompt injection primitive
• Matches in document: 1

Evidence (1 of 1 match):

Line 12:

- [ ] Before installing any skill from ClawdHub

Suggested fix: Clarify the vetting process by specifying what constitutes a reputable source and what steps should be taken to ensure safety before installation.

4. 🟠 SEM-003 — capability_overreach (HIGH)

• Category: Prompt injection / scope override
• Why this matched: The skill suggests reading sensitive configuration files without justification, which could lead to unauthorized access to user credentials.
• Rule intent: Capability claim over-broad relative to the skill's stated purpose
• Matches in document: 1

Evidence (1 of 1 match):

Line 40:

• Reads ~/.ssh, ~/.aws, ~/.config without clear reason

Suggested fix: Remove any instructions that imply accessing sensitive files unless absolutely necessary, and provide clear reasoning for any required access.

5. 🟠 SEM-004 — prompt_injection_subtle (HIGH)

• Category: Prompt injection / scope override
• Why this matched: This instruction could be exploited to manipulate the AI into making decisions that it shouldn't, especially if the user poses as a trusted entity.
• Rule intent: Subtle nudge to hide, mislead, or impersonate that regex couldn't catch
• Matches in document: 1

Evidence (1 of 1 match):

Line 132:

- Ask your human for high-risk decisions

Suggested fix: Rephrase this instruction to clarify that the AI should not act on high-risk decisions without explicit verification of the user's identity and intent.

6. 🟠 SEM-007 — irreversible_action_no_confirmation (HIGH)

• Category: Shell safety
• Why this matched: The directive to reject skills immediately without a confirmation step could lead to accidental loss of data or functionality if misapplied.
• Rule intent: Skill instructs the LLM to take an irreversible action without explicit user confirmation
• Matches in document: 1

Evidence (1 of 1 match):

Line 36:

• 🚨 REJECT IMMEDIATELY IF YOU SEE:

Suggested fix: Introduce a confirmation step before executing any irreversible actions, ensuring that users are aware of the consequences of their decisions.

7. 🟡 SS-004 — sudo_usage (WARNING)

• Category: Shell safety
• Why this matched: Use of sudo for privilege escalation
• Rule intent: Sudo invocation inside the skill body suggests it needs elevated permissions at runtime
• Matches in document: 1

Evidence (1 of 1 match):

Line 48:

     47: • Obfuscated code (compressed, encoded, minified)
>>   48: • Requests elevated/sudo permissions
     49: • Accesses browser cookies/sessions

Suggested fix: Skills should run as a user with the privileges they need. If sudo is required, surface it as a one-time setup step in ## Prerequisites, not in the runtime body.

8. 🔵 QL-001 — shell_block_no_error_handling (INFO)

• Category: quality
• Why this matched: Shell block missing set -e / || exit — silent failures will go unreported
• Rule intent: Shell code blocks without set -e or explicit error handling
• Matches in document: 1

Evidence (1 of 1 match):

Line 109:

    108: For GitHub-hosted skills:
>>  109: ```bash
>>  110: # Check repo stats
>>  111: curl -s "https://api.github.com/repos/OWNER/REPO" | jq '{stars: .stargazers_count, forks: .forks_count, updated: .updated_at}'
>>  112: 
>>  113: # List skill files
>>  114: curl -s "https://api.github.com/repos/OWNER/REPO/contents/skills/SKILL_NAME" | jq '.[].name'
>>  115: 
>>  116: # Fetch and review SKILL.md
>>  117: curl -s "https://raw.githubusercontent.com/OWNER/REPO/main/skills/SKILL_NAME/SKILL.md"
>>  118: ```
    119: 

Suggested fix: Add set -euo pipefail at the top of bash blocks, or chain critical commands with || exit 1. Skills that fail silently mid-script are nearly impossible to debug downstream.

Scope of this edition

The audit covers static rule matching, semantic-layer LLM analysis, and adversarial prompt fuzzing. Three classes of risk live beyond this edition's scope. We name them explicitly:

• Runtime behavior. Verifying what a skill actually does at runtime requires sandboxed execution. That layer ships in a future edition; today's report reflects what the skill states it will do, plus the LLM's read of how it would behave.
• Cross-skill composition. When this skill is chained with others through a planner, the emergent state flow between skills is its own analysis surface. Out of scope for single-skill reports.
• External payloads. A skill that fetches and runs a remote script is flagged at the fetch step. The remote payload itself is audited as a follow-up once the sandbox layer is online.

Methodology

How the score was computed:

1. Document text is scanned against a static rule set of 32 signature patterns. Each rule carries a permanent rule_id (e.g. PI-001), a category, a severity, and a remediation template.
2. Each rule hit deducts from a 100-point base: critical -20, high -10, warning -5, info -1.
3. The letter grade is gated by max severity AND total score: any critical → F; any high → at most D; any warning → at most C; otherwise A/B by score band.
4. Per-category sub-scores apply the same deduction formula to that category's findings only — so you can see WHICH risk surface drove the loss.

Rule matches are augmented by an LLM-based semantic pass when an LLM endpoint is configured. The semantic pass uses rule IDs SEM-001 … SEM-008.

When an LLM endpoint is configured the skill is also probed with a 15-attack adversarial corpus (5 classes × 3 prompts), each judged by a separate LLM call. Failed classes surface as rule IDs AR-001 … AR-005.

Engine + rule set provenance:

• Engine version: 0.2.0
• Rule set version: 1.1.0
• Commit: unknown
• Domain config: general
• Audited at: 2026-07-30T20:42:28.435817Z
• Rules applied: 36 static rules (full registry below)

Known limitations of this report

• False positives are possible. A SKILL.md documenting a dangerous pattern (e.g. an audit skill explaining curl | sh) will match the rule even though the skill's intent is to detect, not execute. Read the matched lines before reacting.
• False negatives are guaranteed in narrow ways. Patterns obfuscated by string concatenation, environment variable indirection, or non-English equivalents will slip past regex.
• Baseline sample size. Same-skill trend analysis (§ Historical baseline) gets meaningful with n≥3 prior audits. With fewer priors the stddev band is widened to avoid false out-of-band signals.

About TAR Engine

TAR Engine is an OSS "wish machine" with built-in audit. Speak a goal; the engine plans, runs and audits skills inside its own container. BYOK. — github.com/qingxuantang/tar-engine