agent-instructions
Use when writing project instructions for a coding agent (CLAUDE.md, AGENTS.md, or equivalent). Covers what belongs in them, what does not, structure, and…
Use when writing an announcement, update, or memo for colleagues. Covers leading with the point, matching the format to the audience, and writing something people actually read.
$ npx -y skills add nimadorostkar/Claude-Skills-collection --skill internal-comms --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/internal-commsContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when writing an announcement, update, or memo for colleagues. Covers leading with the point, matching the format to the audience, and writing something people actually read.
name: internal-comms description: Use when writing an announcement, update, or memo for colleagues. Covers leading with the point, matching the format to the audience, and writing something people actually read. metadata: category: writing version: 1.0.0 tags: [communication, announcement, memo, updates, workplace]
Write internal messages that get read and acted on. The default internal announcement buries its point in paragraph four, and most readers stop at paragraph one.
1. **Write the conclusion first** — What is the one thing the reader must take away? That is the first sentence. Everything after it is support. 2. **State the ask explicitly** — What must the reader do, and by when? A message with an implied ask gets an implied response, which is none. 3. **Structure for skimming** — Bold the key sentence. Use a heading if it is long. Most readers skim; write for that rather than against it. 4. **Cut the preamble** — "As you may know, over the past several quarters the team has been working on..." Delete it. Start at the point. 5. **Match the length to the channel** — A chat message is three sentences. An email is a screen. A memo is a document that people read once, carefully. 6. **Say what happens if they do nothing** — Most readers' real question is whether this affects them.
**Before and after:**
BEFORE — the point is in paragraph three, and nobody reaches it. Hi all, As many of you know, the platform team has been working over the last few months on improving the reliability of our deployment pipeline. This has involved a lot of collaboration across teams and we're grateful for everyone's patience during this process. We've evaluated several approaches, including a full migration to a new CI provider, and after considerable discussion we've decided on an incremental path that we believe balances risk and velocity. Starting next Monday, deploys will require a passing integration test run, which adds approximately 6 minutes to the pipeline. Teams that currently bypass CI with the --force flag will no longer be able to do so. Please update any scripts that use it. ...
AFTER — the point, the ask, and the impact, in that order. **From Monday 21 July, deploys will require a passing integration test run. This adds ~6 minutes to the pipeline, and `--force` will stop working.** **What you need to do:** if any of your scripts use `deploy --force`, remove it before Monday. `grep -r "deploy --force"` across your repo. If you find one and cannot remove it, reply here by Thursday. **Why:** four of the last six production incidents were deploys that bypassed CI. This closes that path. **If you do nothing:** deploys still work, they are 6 minutes slower, and any script using `--force` will fail with a clear error. Details: [RFC-0031](link) | Questions: #platform
Same information. The second is read in fifteen seconds, and the reader knows immediately whether it affects them.
A curated library of 137 production-grade skills for Claude and other AI coding agents. Every skill follows one structure, speaks with one voice, and earns its place by changing what the agent does.
Repo: nimadorostkar/Claude-Skills-collection
Use when writing project instructions for a coding agent (CLAUDE.md, AGENTS.md, or equivalent). Covers what belongs in them, what does not, structure, and…
Use when an agent needs state that survives a session or a context compaction. Covers what to persist, file-based memory, structuring notes for retrieval, and…
Use when automating agent behavior with lifecycle hooks. Covers hook events, deterministic enforcement of rules the model should not be trusted to remember,…
Use when packaging skills, commands, hooks, and MCP servers into a distributable plugin. Covers manifest structure, bundling, versioning, testing, and…
Use when writing a new skill for an AI agent. Covers scoping, description writing for reliable triggering, progressive disclosure, and the difference between a…
Use when reviewing or improving an existing agent skill. Covers triggering accuracy, content quality, redundancy with the base model, and measuring whether the…