Nm Abstract Skill Authoring
v1.8.3Guide to creating Claude Code skills using TDD methodology and persuasion principles. Use for new skill development
⭐ 0· 103·1 current·1 all-time
by@athola
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Benign
high confidencePurpose & Capability
Name and description (skill authoring, TDD, persuasion) align with the content: SKILL.md + modules provide guidance and checklists. The only declared requirement is a config path (night-market.modular-skills), which fits a modular skill authoring workflow; no unrelated credentials or binaries are requested.
Instruction Scope
Runtime instructions are documentation-only (TDD cycles, validators, deployment checklist, persuasive patterns). They reference running local scripts (e.g., python scripts/abstract_validator.py), git tagging, and copying to ~/.claude/skills/, which are normal for an authoring/deployment workflow but the repo/package does not include a scripts/ directory in the provided manifest. The content explicitly teaches strong persuasion/commitment techniques (authoritative 'MUST' language and commitment declarations) to increase compliance — appropriate for its goal but influential, so users should be aware this changes assistant behavior intentionally.
Install Mechanism
No install spec or remote downloads — instruction-only skill carries minimal install risk. There is no archive download or package install configured in this bundle.
Credentials
The skill requires no environment variables or credentials. The single required config path (night-market.modular-skills) is plausible for a modular skill framework. No excessive secrets or unrelated env vars are requested.
Persistence & Privilege
always:false and default autonomous invocation are present (normal). The guide describes copying files into ~/.claude/skills/ and using git for releases — expected for authoring/deployment but these are user actions, not automatic. There is no request to modify other skills' configs or to force permanent presence.
Assessment
This is mainly documentation for building Claude Code skills and appears coherent with that purpose. Things to check before installing or following commands: (1) Inspect the repository/skill package yourself — several commands reference local scripts (scripts/*.py) that are not present in the manifest; don't run scripts you can't review. (2) The guide intentionally teaches strong persuasion/commitment language; if you install this for an agent that can invoke skills autonomously, be aware it will push directive 'MUST' style behaviors. (3) Deployment commands (git tag, git push, cp to ~/.claude/skills/) are user-level operations — confirm remote repo/trust and back up existing skills before copying. (4) If you want to be conservative, test the guide's procedures in an isolated/sandbox environment and verify any referenced validator scripts exist and are safe. (5) If you have concerns about persuasive/influential wording, consider limiting autonomous invocation or reviewing and editing the SKILL.md to soften mandatory language for non-safety-critical contexts.Like a lobster shell, security has layers — review code before you run it.
latestvk97dh7v7jt36wwnz61kzyqnf2n84kmj5
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🛠️ Clawdis
Confignight-market.modular-skills