ClawHub

Argument Selfloop

v1.0.0

Argument self-loop: maintain an argument ledger + premise consistency report for drafted sections. **Trigger**: argument self-loop, argument chain, premise c...

⭐ 0· 110·1 current·1 all-time
by@willoscar
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description describe an argument self-loop and the repository contains Python scripts and pipeline docs that read sections/outline and produce the stated intermediate artifacts (SECTION_ARGUMENT_SUMMARIES.jsonl, ARGUMENT_SKELETON.md, ARGUMENT_SELFLOOP_TODO.md). Required binaries (python3 or python) match the implementation; no unrelated credentials, network, or tools are requested.
ℹ
Instruction Scope
SKILL.md instructs a workflow that reads 'outline/outline.yml' and 'sections/S*.md', performs paragraph-level labeling and produces ledger files. The runtime script (scripts/run.py) implements this read-and-report behavior and writes to workspace/output/*. The SKILL.md also describes applying fixes to section files; the provided run.py does not appear to auto-edit section prose, but an agent following the prose instructions could modify section files. Users should expect the skill to read and write files within the provided workspace.
✓
Install Mechanism
No install spec or network downloads; all code is local Python and there are no external installers or remote artifacts. This is low-risk given the requested runtime (python).
✓
Credentials
The skill declares no environment variables or credentials. It only requires a Python runtime; it does not request unrelated secrets or access to system configuration.
✓
Persistence & Privilege
Flags show always=false and normal model invocation settings. The skill writes only to files inside the provided workspace paths and does not modify other skills or global agent configuration.
Assessment
This skill is coherent with its stated purpose: it reads the repo outline and section Markdown and writes internal ledger files summarizing paragraph-level argument moves. Before installing or running it, confirm you intend an agent to read and write files inside the workspace (sections/* and output/*). Make a repository backup or enable version control so you can review any automated edits; the run.py currently only reads and writes derived output files, but the SKILL.md describes human/agent edits to section files, which an automated agent could perform. There is no network activity or secret access requested, so the main risk is accidental modification of your draft files—review outputs and diffs before merging into your canonical draft.

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

latestvk97bm49awr5yvhn13fms5scscd83671f

License

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

Runtime requirements

Any binpython3, python

Comments