Audit Report: mistral-security-basics — 🟠 D (30/100)

Audited by TAR Engine · 2026-06-28 · 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/jeremylongshore/claude-code-plugins-plus-skills/blob/main/plugins/saas-packs/mistral-pack/skills/mistral-security-basics/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 "mistral-security-basics" skill implements security best practices for Mistral AI integrations, focusing on API key management, prompt injection defense, content moderation, output sanitization, and secure logging. It utilizes tools like secret managers for key management, applies input sanitization techniques to prevent prompt injection, and moderates content using the Mistral Moderation API. Additionally, it automates key rotation and conducts security audits to ensure compliance with established security standards.

Author description: 'Apply Mistral AI security best practices for secrets, prompt injection,

Observed: mistral-security-basics is 7 top-level sections (Overview, Prerequisites, Instructions, Security Audit Checklist, Error Handling, …); ~227 lines of instructions, concise body.

Frontmatter facts:

• Declared allowed-tools: Read, Write, Grep
• Body size: 227 lines / 6452 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)

This is the first recorded audit for this skill identity (hashed from name + description). The baseline section will show mean / stddev / trend after 2+ audits accumulate.

Findings

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

1. 🟠 CE-001 — credential_in_content (HIGH)

• Category: Credential exposure
• Why this matched: Hardcoded API key, secret, or password
• Rule intent: Literal credential value embedded in the skill body (api_key, secret, password, token, etc.)
• Matches in document: 1

Evidence (1 of 1 match):

Line 46:

     45: 
>>   46: # NEVER: api_key = "sk-abc123"
     47: 

Suggested fix: Replace hardcoded secrets with ${VAR_NAME} placeholders and document the env var in SKILL.md ## Required Environment. Rotate any secret that touched git history.

2. 🟠 SEM-006 — credential_handling_unsafe (HIGH)

• Category: Credential exposure
• Why this matched: While the comment advises against logging sensitive information, the lack of enforced measures could lead to accidental exposure of credentials in logs.
• 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 176:

// NEVER log: API keys, message content, user identifiers

Suggested fix: Implement strict logging controls and automated checks to ensure that sensitive information is never logged, reinforcing the comment with actionable code.

3. 🟠 SEM-008 — external_payload_blind_trust (HIGH)

• Category: Malicious payload signatures
• Why this matched: The skill assumes that the output from the moderation API is safe without validating its content, which could lead to processing harmful or inappropriate data.
• 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 106:

const result = await client.classifiers.moderate({

Suggested fix: Add validation checks for the output from the moderation API to ensure it meets safety criteria before further processing.

4. 🟠 SEM-002 — ambiguous_instruction (HIGH)

• Category: Prompt injection / scope override
• Why this matched: The instruction to manage API keys does not specify how to handle potential errors or unauthorized access attempts, which could lead to misuse if a hostile user exploits this ambiguity.
• Rule intent: Ambiguous instruction that could be exploited as a prompt injection primitive
• Matches in document: 1

Evidence (1 of 1 match):

Line 41:

### Step 1: API Key Management

Suggested fix: Clarify the instructions by including specific error handling procedures and security measures to prevent unauthorized access to API keys.

5. 🟠 SEM-004 — prompt_injection_subtle (HIGH)

• Category: Prompt injection / scope override
• Why this matched: The mention of 'XML-wrap' could be misinterpreted or exploited by an attacker to inject malicious XML content, which may not be adequately filtered by the existing sanitization process.
• Rule intent: Subtle nudge to hide, mislead, or impersonate that regex couldn't catch
• Matches in document: 1

Evidence (1 of 1 match):

Line 235:

| Prompt injection | Unsanitized user input | Filter + XML-wrap user content |

Suggested fix: Revise the instruction to specify safe handling of user input and clarify that all user input should be sanitized against various injection vectors, not just filtered.

6. 🟠 SEM-007 — irreversible_action_no_confirmation (HIGH)

• Category: Shell safety
• Why this matched: The automatic rotation of API keys upon authentication failure could lead to unintended consequences if a legitimate user is locked out without confirmation.
• 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 204:

console.error(`Key ${this.current} failed auth, rotating`);

Suggested fix: Implement a confirmation step before rotating keys, ensuring that users are aware of the action and can prevent unnecessary disruptions.

7. 🟡 FA-001 — sensitive_file_access (WARNING)

• Category: Sensitive file access
• Why this matched: Access to sensitive configuration files
• Rule intent: Reads or writes files commonly used to hold secrets (.env, .ssh, .key, .pem)
• Matches in document: 2

Evidence (2 of 2 matches):

Line 102:

    101: 
>>  102: const client = new Mistral({ apiKey: process.env.MISTRAL_API_KEY });
    103: 

Line 220:

    219:         "api_key_from_env": bool(os.environ.get("MISTRAL_API_KEY")),
>>  220:         "gitignore_has_env": ".env" in open(".gitignore").read() if os.path.exists(".gitignore") else False,
    221:         "no_hardcoded_keys": True,  # scan src/ for patterns

Suggested fix: Remove direct references to .env / .ssh / .key / .pem; load secrets from a runtime config service or environment variable instead of naming the file in the skill body.

8. 🟡 SEM-003 — capability_overreach (WARNING)

• Category: Prompt injection / scope override
• Why this matched: The skill grants broad permissions that may not be necessary for its stated purpose, increasing the risk of misuse or unintended actions.
• Rule intent: Capability claim over-broad relative to the skill's stated purpose
• Matches in document: 1

Evidence (1 of 1 match):

Line 15:

allowed-tools: Read, Write, Grep

Suggested fix: Limit the allowed tools to only those necessary for the skill's functionality, reducing the potential attack surface.

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: 2026-06-28T20:34:22.370349Z
• 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