Human Writing

Guidelines and standards for professional, human-like writing and documentation. Use this skill when generating READMEs, technical documentation, code commen...

Security Scan

The skill is an instruction-only style guide for producing human-like technical writing; its declared purpose matches its files and it requests no credentials, installs, or external access.

Purpose & Capability

The name/description (human-like writing guidance) align with the included SKILL.md and two reference documents; nothing in the package requires unrelated capabilities or credentials.

Instruction Scope

Runtime instructions are limited to consulting the bundled reference files, drafting text, and self-auditing for stylistic 'tells'. The skill does not instruct the agent to read arbitrary user files, access environment variables, or transmit data externally.

Install Mechanism

There is no install spec and no code files. As instruction-only content, it writes nothing to disk and does not pull external code or archives.

Credentials

No environment variables, credentials, or config paths are requested; the guidance uses only local bundled reference files, which is proportionate to the stated purpose.

Persistence & Privilege

always is false and the skill does not request special privileges or system-wide changes. Autonomous invocation is allowed by default but is reasonable for a stylistic guidance skill.

Assessment

This skill is an instruction-only style guide and appears internally consistent and low-risk. If you install it, expect the agent to automatically apply these stylistic rules when asked to produce documentation — review outputs to ensure the rules don't remove phrasing you want preserved. Because the skill bundles its reference files and requests no credentials or installs, there is no network or credential exposure from the skill itself. If you have organizational style rules or need different tone choices for some outputs, keep that in mind and test the skill on representative prompts before wide use.

Human-Writing Skill

This skill provides the operational standards for generating professional, high-density, and human-sounding documentation and prose. It is designed to purge common LLM stylistic "tells" and replace them with the precision of a senior engineer or domain expert.

Core Directives

1. Eliminate AI "Tells": Before finalizing any documentation or formal text, refer to ai-tells.md to identify and remove overused LLM vocabulary, structural tropes, and puffy language.
2. Apply Professional Standards: Follow the guidelines in standards.md for technical precision, information density, and tone.
3. No Buzzwords: Zero tolerance for "synergy," "cutting-edge," "revolutionize," "seamless," or "leverage." If a technical term exists, use it.
4. No Emojis in Docs: Reserve emojis for chat interactions (as per SOUL.md). Professional documentation (READMEs, PR descriptions, code comments) must remain text-only for maximum clarity.
5. Precision Over Prose: Humans in the field value numbers, versions, and RFCs over flowery descriptions.

Workflow

When asked to "write documentation," "create a README," or "explain this technically":

1. Scan references/ai-tells.md for words to ban from the current draft.
2. Apply references/standards.md to structure the output with high density and low fluff.
3. Draft the content.
4. Self-Audit: Verify the output does not contain "Rule of Three" adjectives or "Not only... but also" parallelisms.
5. Finalize: Remove all emojis and corporate filler.

Reference Materials

• ai-tells.md - Field guide to AI writing "tells" to avoid.
• standards.md - Human-like professional writing standards.