Audit Report: ai-analysis — 🟠 D (40/100)
Audited by TAR Engine · 2026-07-26 · 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.
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 ai-analysis skill facilitates the comparison of two or more AI models by enforcing strict experimental parity across model tiers, sampling configurations, and cost transparency. It blocks execution until all necessary parity checks are verified or explicitly waived by the user, ensuring fair and transparent benchmarking. The skill logs all API calls made during the process to maintain a complete record of the comparisons.
Author description: Use whenever the user is doing research, benchmarks, or comparisons that involve TWO OR MORE AI models, agents, or LLMs (closed or open source). Enforces experimental parity across model tier, sampling configuration (temperature, max_tokens, reasoning effort, system prompt, seed), and cost transparency. BLOCKS execution until parity is verified or the user has explicitly approved an asymmetry. Activates on phrases like "compare GPT-4 vs Claude", "benchmark these models", "evaluate Llama against Gemini", "which model is better at X", or any research workflow that produces side-by-side AI outputs.
Observed: ai-analysis is 6 top-level sections (The Five Parity Checks, Raw call logging (mandatory for every API call), Cost logging (after each run), What to do when the user pushes back, Final report format, …); ~219 lines of instructions, makes outbound network calls, concise body.
Frontmatter facts:
- Body size: 219 lines / 14816 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.
| Category | Rules evaluated | Findings | Max severity | Sub-score |
|---|---|---|---|---|
| Prompt injection / scope override | 5 | 3 | 🟠 high | 70/100 |
| Shell safety | 4 | 1 | 🟠 high | 90/100 |
| Sensitive file access | 1 | 0 | ⚪ none | 100/100 |
| Data exfiltration | 3 | 0 | ⚪ none | 100/100 |
| Credential exposure | 1 | 1 | 🟠 high | 90/100 |
| Malicious payload signatures | 3 | 1 | 🟠 high | 90/100 |
| Supply chain (deps + CVE) | 0 | 0 | ⚪ none | 100/100 |
| quality | 2 | 0 | ⚪ none | 100/100 |
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
6 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: This instruction could lead to sensitive information being logged or exposed if not handled properly, as it does not specify how to manage other potentially sensitive 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 173:
Redact only credentials.
Suggested fix: Expand the instruction to include guidelines on handling other sensitive data, such as user prompts or proprietary information, to prevent accidental exposure.
2. 🟠 SEM-008 — external_payload_blind_trust (HIGH)
- Category: Malicious payload signatures
- Why this matched: The skill relies on external sources for pricing without validating the integrity of the fetched data, which could lead to using outdated or incorrect pricing information.
- 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 66:
Always pull the latest pricing fresh via `WebSearch` against the official pricing page for each provider.
Suggested fix: Implement a validation step to check the integrity and accuracy of the fetched pricing data before using it in calculations or comparisons.
3. 🟠 SEM-002 — ambiguous_instruction (HIGH)
- Category: Prompt injection / scope override
- Why this matched: This instruction could lead to a situation where the skill makes assumptions about parameters that the user did not intend, potentially leading to unfair comparisons.
- Rule intent: Ambiguous instruction that could be exploited as a prompt injection primitive
- Matches in document: 1
Evidence (1 of 1 match):
Line 41:
If the user has not specified values, do NOT pick defaults silently.
Suggested fix: Clarify that the skill should always prompt the user for confirmation on default values before proceeding, ensuring that user intent is clear and respected.
4. 🟠 SEM-004 — prompt_injection_subtle (HIGH)
- Category: Prompt injection / scope override
- Why this matched: This instruction could be exploited by a malicious user to manipulate the logging of sensitive content, leading to potential data leaks.
- Rule intent: Subtle nudge to hide, mislead, or impersonate that regex couldn't catch
- Matches in document: 1
Evidence (1 of 1 match):
Line 184:
For genuinely sensitive content (PII, proprietary text, unpublished research) the user may want stronger handling.
Suggested fix: Implement stricter controls and validation for logging sensitive content, ensuring that any user input is sanitized and that the skill does not log sensitive information without explicit user consent.
5. 🟠 SEM-003 — capability_overreach (HIGH)
- Category: Prompt injection / scope override
- Why this matched: The requirement to log every API call in full fidelity may expose sensitive information unnecessarily and could lead to compliance issues.
- Rule intent: Capability claim over-broad relative to the skill's stated purpose
- Matches in document: 1
Evidence (1 of 1 match):
Line 96:
Every call to any LLM made during this skill... MUST be saved to disk in full fidelity.
Suggested fix: Consider limiting the logging to essential information only and provide options for users to opt-out of logging sensitive data.
6. 🟠 SEM-007 — irreversible_action_no_confirmation (HIGH)
- Category: Shell safety
- Why this matched: While this instruction does require user confirmation, it does not specify that the user must confirm the total cost before proceeding, which could lead to unexpected charges.
- 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 84:
Get an explicit 'yes proceed' before any paid API call.
Suggested fix: Ensure that the skill explicitly states the total estimated cost and requires user confirmation on that total before making any paid API calls.
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:
- 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. - Each rule hit deducts from a 100-point base: critical -20, high -10, warning -5, info -1.
- 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.
- 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:
2026-07-26T20:43:41.284669Z - Rules applied: 36 static rules (full registry below)
Full rule registry applied to this audit
| Rule ID | Name | Category | Severity | |---|---|---|:---:| | `FA-001` | sensitive_file_access | file_access | warning | | `SS-001` | destructive_bash | shell_safety | high | | `SS-002` | force_flag_abuse | shell_safety | high | | `DE-001` | external_data_exfil | data_exfil | high | | `CE-001` | credential_in_content | credential_exposure | high | | `SS-003` | pipe_to_shell | shell_safety | critical | | `SS-004` | sudo_usage | shell_safety | warning | | `PI-001` | prompt_injection_bypass | prompt_injection | critical | | `PI-002` | hidden_instruction | prompt_injection | critical | | `PI-003` | excessive_permission_claim | prompt_injection | high | | `PI-004` | disable_safety_instruction | prompt_injection | high | | `PI-005` | impersonation_instruction | prompt_injection | high | | `MP-001` | encoded_payload | malicious_payload | warning | | `DE-002` | network_exfil_pattern | data_exfil | high | | `MP-002` | crypto_miner_pattern | malicious_payload | critical | | `MP-003` | reverse_shell_pattern | malicious_payload | critical | | `DE-003` | data_collection_broad | data_exfil | warning | | `QL-001` | shell_block_no_error_handling | quality | info | | `QL-002` | unpinned_install_command | quality | info | | `SEM-001` | semantic_evasion | prompt_injection | high | | `SEM-002` | ambiguous_instruction | prompt_injection | warning | | `SEM-003` | capability_overreach | prompt_injection | warning | | `SEM-004` | prompt_injection_subtle | prompt_injection | high | | `SEM-005` | unauthorized_data_flow | data_exfil | high | | `SEM-006` | credential_handling_unsafe | credential_exposure | high | | `SEM-007` | irreversible_action_no_confirmation | shell_safety | high | | `SEM-008` | external_payload_blind_trust | malicious_payload | high | | `AR-001` | instruction_override_failure | prompt_injection | high | | `AR-002` | role_jailbreak_failure | prompt_injection | high | | `AR-003` | hidden_payload_failure | malicious_payload | high | | `AR-004` | authority_spoof_failure | prompt_injection | high | | `AR-005` | reflective_injection_failure | prompt_injection | high | | `SUP-001` | typosquat_risk | supply_chain | high | | `SUP-002` | known_vulnerability | supply_chain | high | | `SUP-003` | unpinned_dependency | supply_chain | warning | | `SUP-004` | deprecated_or_yanked | supply_chain | warning |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