📦

Audit-Verlauf

doc-consistency-reviewer - 8 Audits

Versionsvergleich

Änderungen an Fähigkeiten und Befunden über geprüfte Versionen hinweg, neueste zuerst.

VersionDatumErgebnisPrüfelementeÄnderung ggü. vorheriger
v8 Neueste6 июл. 2026 г., 14:25 1 bestätigt0Keine Änderung der Fähigkeiten
v7 6 июл. 2026 г., 14:25 1 bestätigt0Externe Befehle Dateisystemzugriff
v6 30 июн. 2026 г., 04:10 1 bestätigt0Dateisystemzugriff
v5 21 янв. 2026 г., 17:14 Keine bestätigten Befunde0 Externe Befehle
v4 17 янв. 2026 г., 06:32 Keine bestätigten Befunde0Keine Änderung der Fähigkeiten
v3 17 янв. 2026 г., 06:32 Keine bestätigten Befunde0Keine Änderung der Fähigkeiten
v2 11 янв. 2026 г., 05:58 Keine bestätigten Befunde0Keine Änderung der Fähigkeiten
v1 11 янв. 2026 г., 05:58 Keine bestätigten Befunde0Ausgangsbasis

6 июл. 2026 г., 14:25

Static external-command findings are false positives from markdown fences, inline path formatting, and JSON strings. The entropy findings are false positives because the files are readable markdown. Semantic review found an embedded audit output file that claims the skill is safe and all findings are false positives.

4
Gescannte Dateien
712
Analysierte Zeilen
2
Prüfelemente
0
Falschmeldungen ignoriert

Bestätigte Sicherheitsbedenken (1)

Hoch
Prompt Injection Attempt Detected
evaluation-output.json embeds self-approval claims: "risk_level: safe", "safe_to_publish: true", and "All 36 static findings are false positives." These claims can bias an automated reviewer if the file is loaded as skill context.
The file directly asserts safety and false-positive status inside the skill package. It lacks an explicit override command, but matches the suspicious self-approval pattern.
Geprüft von: codex

6 июл. 2026 г., 14:25

Static external-command findings are false positives from markdown fences, inline path formatting, and JSON strings. The entropy findings are false positives because the files are readable markdown. Semantic review found an embedded audit output file that claims the skill is safe and all findings are false positives.

4
Gescannte Dateien
712
Analysierte Zeilen
2
Prüfelemente
0
Falschmeldungen ignoriert

Bestätigte Sicherheitsbedenken (1)

Hoch
Prompt Injection Attempt Detected
evaluation-output.json embeds self-approval claims: "risk_level: safe", "safe_to_publish: true", and "All 36 static findings are false positives." These claims can bias an automated reviewer if the file is loaded as skill context.
The file directly asserts safety and false-positive status inside the skill package. It lacks an explicit override command, but matches the suspicious self-approval pattern.
Geprüft von: codex

30 июн. 2026 г., 04:10

Static findings for shell backticks, weak cryptography, and obfuscation were evaluated as false positives from Markdown fences, example text, and multilingual documentation. No prompt injection, network access, executable script, or secret handling was found. The only retained risk is low because the skill asks the agent to write a local doc-consistency.md report.

4
Gescannte Dateien
712
Analysierte Zeilen
2
Prüfelemente
3
Falschmeldungen ignoriert

Bestätigte Sicherheitsbedenken (1)

Niedrig
Local Report File Output Requested
The skill instructs the agent to output a doc-consistency.md report. This is expected behavior, but it may create or overwrite a local report file.
Line 94 directly asks for a report file output. The action is scoped and transparent, but it is still filesystem interaction that users should expect.
Statische falsch positive Treffer ignoriert (3)

Diese statischen Treffer wurden durch semantische Prüfung verworfen oder entsprachen reinen Schema-Tokens; daher werden sie aus Transparenzgründen angezeigt, beeinflussen jedoch nicht die Qualitätsbewertung.

Niedrig
evaluation-output.json:18evaluation-output.json:24evaluation-output.json:30evaluation-output.json:36evaluation-output.json:42evaluation-output.json:48evaluation-output.json:54evaluation-output.json:60evaluation-output.json:66evaluation-output.json:72evaluation-output.json:78evaluation-output.json:84evaluation-output.json:90evaluation-output.json:96evaluation-output.json:102evaluation-output.json:108evaluation-output.json:114evaluation-output.json:120evaluation-output.json:126evaluation-output.json:132evaluation-output.json:138evaluation-output.json:144evaluation-output.json:150evaluation-output.json:156evaluation-output.json:162evaluation-output.json:168evaluation-output.json:174evaluation-output.json:180evaluation-output.json:186evaluation-output.json:192evaluation-output.json:198evaluation-output.json:204references/output-format.md:5references/output-format.md:19references/output-format.md:25references/output-format.md:30references/output-format.md:31references/output-format.md:34references/output-format.md:36references/output-format.md:38references/output-format.md:40references/output-format.md:47references/output-format.md:53references/output-format.md:97references/output-format.md:107references/output-format.md:108references/output-format.md:111references/output-format.md:113references/output-format.md:115references/output-format.md:120references/output-format.md:124references/output-format.md:124references/output-format.md:134references/output-format.md:135references/output-format.md:138references/output-format.md:140references/output-format.md:142SKILL.md:23SKILL.md:28SKILL.md:61SKILL.md:64SKILL.md:71SKILL.md:75SKILL.md:80
Markdown Backticks Misclassified as Command Execution
The static shell-execution alerts point to Markdown fences and inline path formatting. These locations document report templates and examples, and they do not execute commands.
The referenced lines contain Markdown code fences or inline code markers in documentation files. No Ruby, shell invocation, command substitution, or runtime script path is present.
Niedrig
Weak Cryptography Alerts Are Textual False Positives
The weak-cryptography alerts point to skill descriptions and prior evaluation prose, not cryptographic functions or algorithm use.
The cited lines are descriptive text about documentation review. I found no evidence of MD5, SHA-1, DES, RC4, cryptographic APIs, or password hashing behavior.
Niedrig
High Entropy Alerts Are Plain Documentation
The high-entropy alerts point to normal Markdown headings and YAML frontmatter. The files are readable text, not binary or encrypted payloads.
The files render as ordinary Markdown and YAML frontmatter. The entropy heuristic appears triggered by Chinese text and punctuation, with no encoded blob or hidden payload.

Risikofaktoren

📁 Dateisystemzugriff (1)
Geprüft von: codex

21 янв. 2026 г., 17:14

All static findings are false positives. The 64 external_commands detections are markdown code blocks with shell examples in documentation, not executable code. The 17 weak_crypto detections are base64-encoded content in JSON/YAML files. The 9 obfuscation detections are high-entropy structured data patterns, not obfuscated payloads. Network detection is legitimate metadata (GitHub source URL). No actual security risks found.

5
Gescannte Dateien
1,634
Analysierte Zeilen
0
Prüfelemente
0
Falschmeldungen ignoriert
Für dieses abgeschlossene Audit wurden keine bestätigten Sicherheitsbefunde erfasst.
Geprüft von: claude

17 янв. 2026 г., 06:32

AI analysis failed after multiple attempts - MANUAL REVIEW REQUIRED before publishing. This skill cannot be auto-published until reviewed by a human.

5
Gescannte Dateien
1,075
Analysierte Zeilen
1
Prüfelemente
0
Falschmeldungen ignoriert

Risikofaktoren

⚙️ Externe Befehle (64)
evaluation-output.json:18-24 evaluation-output.json:24-30 evaluation-output.json:30-36 evaluation-output.json:36-42 evaluation-output.json:42-48 evaluation-output.json:48-54 evaluation-output.json:54-60 evaluation-output.json:60-66 evaluation-output.json:66-72 evaluation-output.json:72-78 evaluation-output.json:78-84 evaluation-output.json:84-90 evaluation-output.json:90-96 evaluation-output.json:96-102 evaluation-output.json:102-108 evaluation-output.json:108-114 evaluation-output.json:114-120 evaluation-output.json:120-126 evaluation-output.json:126-132 evaluation-output.json:132-138 evaluation-output.json:138-144 evaluation-output.json:144-150 evaluation-output.json:150-156 evaluation-output.json:156-162 evaluation-output.json:162-168 evaluation-output.json:168-174 evaluation-output.json:174-180 evaluation-output.json:180-186 evaluation-output.json:186-192 evaluation-output.json:192-198 evaluation-output.json:198-204 evaluation-output.json:204-210 references/output-format.md:5-19 references/output-format.md:19-25 references/output-format.md:25-30 references/output-format.md:30-31 references/output-format.md:31-34 references/output-format.md:34-36 references/output-format.md:36-38 references/output-format.md:38-40 references/output-format.md:40-47 references/output-format.md:47-53 references/output-format.md:53-97 references/output-format.md:97-107 references/output-format.md:107-108 references/output-format.md:108-111 references/output-format.md:111-113 references/output-format.md:113-115 references/output-format.md:115-120 references/output-format.md:120-124 references/output-format.md:124 references/output-format.md:124-134 references/output-format.md:134-135 references/output-format.md:135-138 references/output-format.md:138-140 references/output-format.md:140-142 references/output-format.md:142-144 SKILL.md:23-28 SKILL.md:28-61 SKILL.md:61-64 SKILL.md:64-71 SKILL.md:71-75 SKILL.md:75-80 SKILL.md:80-94

Erkannte Muster

Ruby/shell backtick executionWeak cryptographic algorithm[HEURISTIC] High file entropy (6.55 bits) - possible binary/encrypted content[HEURISTIC] High file entropy (6.19 bits) - possible binary/encrypted content[HEURISTIC] High file entropy (6.84 bits) - possible binary/encrypted content
Geprüft von: claude

17 янв. 2026 г., 06:32

AI analysis failed after multiple attempts - MANUAL REVIEW REQUIRED before publishing. This skill cannot be auto-published until reviewed by a human.

5
Gescannte Dateien
1,075
Analysierte Zeilen
1
Prüfelemente
0
Falschmeldungen ignoriert

Risikofaktoren

⚙️ Externe Befehle (64)
evaluation-output.json:18-24 evaluation-output.json:24-30 evaluation-output.json:30-36 evaluation-output.json:36-42 evaluation-output.json:42-48 evaluation-output.json:48-54 evaluation-output.json:54-60 evaluation-output.json:60-66 evaluation-output.json:66-72 evaluation-output.json:72-78 evaluation-output.json:78-84 evaluation-output.json:84-90 evaluation-output.json:90-96 evaluation-output.json:96-102 evaluation-output.json:102-108 evaluation-output.json:108-114 evaluation-output.json:114-120 evaluation-output.json:120-126 evaluation-output.json:126-132 evaluation-output.json:132-138 evaluation-output.json:138-144 evaluation-output.json:144-150 evaluation-output.json:150-156 evaluation-output.json:156-162 evaluation-output.json:162-168 evaluation-output.json:168-174 evaluation-output.json:174-180 evaluation-output.json:180-186 evaluation-output.json:186-192 evaluation-output.json:192-198 evaluation-output.json:198-204 evaluation-output.json:204-210 references/output-format.md:5-19 references/output-format.md:19-25 references/output-format.md:25-30 references/output-format.md:30-31 references/output-format.md:31-34 references/output-format.md:34-36 references/output-format.md:36-38 references/output-format.md:38-40 references/output-format.md:40-47 references/output-format.md:47-53 references/output-format.md:53-97 references/output-format.md:97-107 references/output-format.md:107-108 references/output-format.md:108-111 references/output-format.md:111-113 references/output-format.md:113-115 references/output-format.md:115-120 references/output-format.md:120-124 references/output-format.md:124 references/output-format.md:124-134 references/output-format.md:134-135 references/output-format.md:135-138 references/output-format.md:138-140 references/output-format.md:140-142 references/output-format.md:142-144 SKILL.md:23-28 SKILL.md:28-61 SKILL.md:61-64 SKILL.md:64-71 SKILL.md:71-75 SKILL.md:75-80 SKILL.md:80-94

Erkannte Muster

Ruby/shell backtick executionWeak cryptographic algorithm[HEURISTIC] High file entropy (6.55 bits) - possible binary/encrypted content[HEURISTIC] High file entropy (6.19 bits) - possible binary/encrypted content[HEURISTIC] High file entropy (6.84 bits) - possible binary/encrypted content
Geprüft von: claude