ClawHub

Outline Revision Annotator

v1.1.0

Use when comparing a revised PDF outline, script, treatment, beat sheet, season map, or character bio against an older DOCX/PDF version and you need a high-s...

⭐ 0· 56·0 current·0 all-time
by@hshgogogo
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
✓
Purpose & Capability
Name/description, SKILL.md, and the included script all align: they extract text from a new PDF and an old source, accept a human-authored manifest of structural changes, and render an annotated PDF + markdown/DOCX report. Required dependencies (PyMuPDF, python-docx) match the stated functionality.
ℹ
Instruction Scope
The SKILL.md keeps the agent responsible for judgment (what counts as a structural change) and restricts the script to extraction and rendering. The auto-follow hook allows the skill to run after the agent edits documents — this is coherent for an auto-reporting workflow but means the agent may trigger generation without a separate user prompt when its own edits meet the listed conditions.
✓
Install Mechanism
No automatic install spec is present (instruction-only). Installation instructions recommend pip-installing well-known Python packages (pymupdf, python-docx). That is a low-risk, proportionate approach for the task.
✓
Credentials
The skill requests no environment variables, no credentials, and no config paths. All file access is local and limited to the user-supplied PDF/docx paths and output folders described in the workflow.
ℹ
Persistence & Privilege
always:false (normal). The combination of the skill's auto-follow hook and the platform-default ability for the agent to invoke skills autonomously means the skill can run automatically after the agent edits relevant documents. This is reasonable for the use case but users should be aware it can generate outputs without an additional explicit user command when the hook conditions are met.
Assessment
This skill appears coherent with its stated purpose, but before installing or running it: (1) inspect the full script locally (especially if you were shown a truncated snippet) to confirm there are no hidden network calls or unexpected file writes; (2) run it in a benign working folder (not on sensitive system directories) and review generated annotation_manifest.json before rendering the PDF; (3) install Python deps from PyPI only if you trust those packages and consider using a virtualenv; (4) be aware the auto-follow hook can trigger automatic generation after the agent edits documents — opt out or disable autonomous invocation if you do not want automatic runs; (5) verify CJK fonts are available on your system if you expect Chinese output.

Like a lobster shell, security has layers — review code before you run it.

latestvk974raqkmepw9hsp4achx0edv984t014

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments