Install
$ agentstack add skill-wolido-subagent-isolation-writing-clearly-and-concisely ✓ 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
Writing Clearly and Concisely
Overview
Three-level optimizer: clarity rules, AI patterns, voice injection. Scan Quick Reference first; load references as needed.
Important
- Preserve meaning. Never change facts or technical details.
- Preserve tone intent. Match desired tone. No casual→formal unless asked.
- Don't over-edit. Leave clean sentences unchanged. Single-sentence
changes should not exceed ~30% unless clearly broken.
- Level 3 is context-sensitive. Inject personality only in personal
blogs, social media, informal emails, or when user asks for "more human". Skip for technical docs, legal text, error messages, and UI copy.
When to Use / When NOT to Use
Use: docs, READMEs, commits, PRs, error messages, UI copy, reports, summaries, puffy/robotic/AI-generated prose. Do NOT use: fiction/poetry, legal contracts, translation-only, code, data analysis, math, tone-shifting without clarity needs.
Quick Reference
Level 1 — Foundation: Load 03-elementary-principles-of-composition.md for most tasks. Load 02-... for grammar issues, 05-... for word-choice debates.
Level 2 — Detection: Scan first, load references/signs-of-ai-writing.md for details. Core: C1–C7 Content, L1–L5 Language, S1–S5 Style, CM1–CM3 Communication, F1–F3 Filler/Hedging. Extended: M1–M4 Markup, H1–H4 Historical (check if context permits).
Level 3 — Enhancement (Inject voice): Use ONLY for personal/informal contexts. Skip for technical/legal/UI text.
- Opinions over neutral lists; vary sentence rhythm
- Acknowledge complexity; use "I" when appropriate
- Be specific about feelings ("unsettling" > "concerning")
Process
If vague ("polish", "润色"), default conservative: Level 1 + 2. Do NOT apply Level 3 unless promotional/personal or user asks for "more human".
- Read input. Note context.
- Level 1 — Load
03-.... Check active voice, concision,
parallel structure. Skip if clean.
- Level 2 — Scan C1–F3. If suspicious, load
signs-of-ai-writing.md
for guidance. M1–M4: check only if text contains Markdown/wikitext markup or is intended for Wikipedia/MediaWiki.
- Level 3 — ONLY if personal/informal or user asks for "more human".
Skip for technical docs, legal, errors, UI, reports.
- Quality check — Natural aloud, varied structure, specifics replace
vagueness. If <3 minor changes, return original with note.
- Output — Preserve markdown. Return rewritten text + bullet list
of changes. See references/full-example.md for format.
Common Issues
See references/common-issues.md for typical pitfalls and how to avoid them.
Reference
references/signs-of-ai-writing.md— AI patterns C1–H4.references/03-elementary-principles-of-composition.md— Strunk Rules 8–18.references/02-elementary-rules-of-usage.md— Grammar Rules 1–7.references/04-a-few-matters-of-form.md— Formatting.references/05-words-and-expressions-commonly-misused.md— Word choice.references/full-example.md— Worked example.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: Wolido
- Source: Wolido/subagent-isolation
- License: MIT
- Homepage: https://www.npmjs.com/package/subagent-isolation
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.