Install
$ agentstack add skill-orkas-ai-orkas-awesome-agentskills-content-writer ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README — it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming — see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps — measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Content Writer
When to Use
Use this skill when the user wants to write, improve, or audit long-form content.
Use for:
- writing articles, blogs, newsletters, tutorials, case studies, explainers, or opinion pieces.
- building outlines, hooks, section plans, titles, conclusions, and citation lists.
- improving an existing draft while preserving the author's voice.
- making text more natural and less AI-like.
- checking facts, statistics, citations, links, and company/product claims before publishing.
Do not use this skill for platform-native social posts, social calendars, browser-based draft publishing, or final professional legal/medical/financial advice.
Choose The Mode
Pick one mode from the user's request. If the request spans multiple modes, handle them in order and say which mode is active.
| Mode | Use when | Output | |---|---|---| | plan | user needs topic clarification, angle, outline, or research plan | thesis, reader, outline, evidence gaps | | draft | user wants a first complete article or section | full draft with assumptions and source gaps | | revise | user has a draft and wants structure, clarity, or style edits | edited version plus change notes | | humanize | user says it sounds AI-like, stiff, generic, or unnatural | natural rewrite preserving meaning | | audit | user asks to verify facts, links, references, claims, or publish readiness | audit report with pass/issues |
Workflow
- Clarify only what matters:
- topic, target reader, publishing context, purpose, tone, length, and deadline.
- available materials: notes, draft, links, files, citations, brand voice, product docs, or style samples.
- whether the user wants speed, depth, or strict verification.
- For
plan:
- define thesis, reader promise, angle, hook options, section structure, and evidence needs.
- mark weak claims and missing sources.
- For
draft:
- write from the outline or available source material.
- use clear labels for assumptions, placeholders, and claims needing verification.
- avoid inventing statistics, experts, sources, case studies, customer quotes, or publication dates.
- For
revise:
- improve clarity, logic, flow, evidence, section order, rhythm, and conclusion.
- preserve the user's intended meaning and voice.
- give optional alternatives when tone is subjective.
- For
humanize:
- remove inflated significance, promotional phrasing, vague attribution, generic transitions, forced groups of three, excessive synonym cycling, and uniform sentence rhythm.
- prefer concrete verbs, specific nouns, natural uncertainty, and human pacing.
- do not make unsupported claims more confident.
- For
audit:
- extract URLs, statistics, research citations, company claims, source attributions, and competitor claims.
- verify links and exact claim support when tools/web access are available.
- label findings:
PASS,BROKEN,FETCH_FAILED,ACCESS_BLOCKED,MISMATCH,UNSOURCED,UNVERIFIABLE, orUNVERIFIABLE_COMPANY_CLAIM.
- Finish with the most useful next action:
- draft, revised text, checklist, verification gaps, or recommended publication hold.
Return Formats
For planning:
# Content Plan
## Thesis
- ...
## Reader
- ...
## Hook Options
1. ...
2. ...
3. ...
## Outline
### 1. [section]
- Point:
- Evidence needed:
- Example:
## Source Gaps
- ...
For revision or humanizing:
## Rewritten Version
...
## What Changed
- ...
## Still Needs Verification
- ...
For audit:
# Content Audit
## Summary
| Category | Total | Pass | Issues |
|---|---:|---:|---:|
## Must Fix Before Publishing
| Claim | Status | Issue | Suggested Fix |
|---|---|---|---|
## Warnings
- ...
## Passed / Limited
- ...
Dependencies
- Web access is required for current research, source verification, and live URL checks.
- User-provided drafts, source files, brand claims, product docs, and style samples take priority over generic assumptions.
- Some sources may require login or manual access; mark them as access-limited.
Limits
- Do not fabricate facts, data, citations, links, quotes, customer stories, or personal experience.
- Do not guarantee that text passes AI detection.
- Do not remove required disclosures, citations, compliance language, or safety warnings.
- Do not publish or operate external platforms.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Orkas-AI
- Source: Orkas-AI/Orkas-Awesome-AgentSkills
- License: MIT
- Homepage: https://orkas.ai
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.