Agent Economy ReportAgent Economy ReportSAFE version 1.0.2 · audited 2026-09-17
This package is a Word/DOCX skill made of Markdown guidance only. It explains how to create, inspect, and edit DOCX files while preserving formatting, tracked changes, tables, sections, and compatibility. No executable code, hidden network calls, credential access, or persistence were found in the provided files. The main user-facing risks are sensitive hidden document content and malformed OOXML edits, both of which the package itself warns about. The only command-like lines are ClawHub feedback or install suggestions and are not hidden or automatic.
| Declared purpose | Create, inspect, and edit Microsoft Word documents and DOCX files with reliable styles, numbering, tracked changes, tables, sections, and compatibility checks. |
| Observed behavior | The package contains only Markdown instructions and metadata. It tells an agent how to treat DOCX files as OOXML ZIP packages, preserve styles and numbering, handle tracked changes and comments, and verify round-trip compatibility. There is no executable code, no network request logic, no file-write implementation, and no persistence mechanism. Command-like lines are limited to user-facing ClawHub install, star, and sync suggestions. |
| Verdict, rules only | SAFE |
| Verdict, AI | SAFE |
| Final verdict | SAFE (never better than either pass) |
| capability | vs purpose | detail |
|---|---|---|
| filesystem_read | needed | The declared purpose requires reading and inspecting DOCX files and their XML parts. |
| filesystem_write | needed | The declared purpose requires creating or editing Word documents and DOCX files. |
| shell | exceeds purpose | No shell implementation is present; only user-facing ClawHub command suggestions appear at SKILL.md lines 96, 103, and 104. |
| network | exceeds purpose | Only external homepage and reference URLs appear in metadata, with no network call behavior implemented. |
| severity | finding |
|---|---|
| info | Declared purpose This is the package's stated purpose and matches the observed Markdown guidance. SKILL.md:6 Create, inspect, and edit Microsoft Word documents and DOCX files with reliable styles, numbering, tracked changes, tables, sections, and compatibility checks. |
| info | False positive: 'node' is an XML node, not the Node.js runtime The deterministic facts list a 'node' binary on this line, but the text refers to an OOXML XML node. No executable binary or shell call is present. SKILL.md:21 Text may be split across multiple runs; never assume one word or sentence lives in one XML node. |
| info | Hidden document content warning The package warns that deleted or hidden text may remain in document XML. This is relevant for sensitive documents but is not malicious behavior. SKILL.md:57 Deleted text may still exist in the XML even when it no longer appears on screen. |
| info | Macro-bearing file risk noted The guidance distinguishes macro-bearing DOCM and legacy DOC files as higher risk and advises caution. SKILL.md:74 Treat `.docm` as macro-bearing and higher risk; treat `.doc` as legacy input that may need conversion first. |
| info | Sensitive content risk disclosed The package tells users that document internals may contain sensitive or hidden content, an appropriate risk disclosure. skill-card.md:25 Risk: Word documents can contain tracked changes, comments, deleted text, legal text, or sensitive business content that may still exist in document internals. |
| info | Macro execution avoided The package explicitly advises against executing macros and recommends inspection before edits. skill-card.md:35 Mitigation: Treat macro-bearing and legacy files as higher risk, avoid executing macros, and convert or inspect them with appropriate tooling before making DOCX |
| low | Related skill install suggestion This line suggests installing related skills but only if the user confirms. It is not automatic execution and does not exceed the package purpose. SKILL.md:96 Install with `clawhub install <slug>` if user confirms: |
| info | Feedback command A visible ClawHub feedback command. No automatic or hidden execution is present. SKILL.md:103 - If useful: `clawhub star word-docx` |
| info | Sync command A visible ClawHub sync suggestion. It does not perform hidden actions or access credentials. SKILL.md:104 - Stay updated: `clawhub sync` |
0 AI finding(s) were dropped because their file, line or quote did not match the package.
| domain | service rating | where |
|---|---|---|
| clawhub.ai | not in the service index | skill-card.md:9 |
| clawic.com | not in the service index | SKILL.md:5 |
| Binaries invoked | node |
| Environment variables read | none |
| Hard-coded wallets | none |
| Pipes a download to a shell | 0 |
| eval / exec / subprocess | 0 |
| base64 blobs | 0 |
| File writes | 0 |
| Persistence | 0 |
| file | lines | sha256 |
|---|---|---|
SKILL.md | 104 | c766da489810acf0… |
_meta.json | 6 | bf173c91d79a5487… |
skill-card.md | 58 | ee543d852ff1ee29… |
For agents
JSON: https://agenteconomy.report/k/word-docx.audit.json · badge: https://agenteconomy.report/k/word-docx.audited.svg ·
skill rating: /k/word-docx · commission an audit of any skill: US$ 29 per version.
The complete published package of this exact version was downloaded from the registry and read statically; nothing was executed. A deterministic pass extracts network destinations, binaries, environment variables, writes, install commands, obfuscation markers, subprocesses and wallets, each with file and line. An AI then reads the whole package with those facts and writes the summary, the capabilities and the findings under a strict schema; every finding it produces must cite a file, a line and the exact text, or it is dropped. The final verdict is the worst of the two passes. The audit does not change the skill's trust tier (policy); the author may respond through the dispute channel and the response is published here. Commissioned by: the Agent Economy Report (free program: the 150 most downloaded skills, September 2026).