Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Gnet Reason To Write

ASecurity

Picks one reason to write per person that is useful to them, why it fits, and what not to ask yet. Use for "run gnet-reason-to-write", "every message I start is an ask", "what do I say to them", "reason to message someone", "how to reach out without asking for a job", "I don't want to sound needy", "reconnect with an old contact", "why would they reply to me", part of the Claude Graduate Networking Guide Pack by Polar Bear.

2 stars
0 votes
0 copies
0 views
Added 10/5/2026
ai-agentsgo

Works with

cli

Security Analysis

A100/100

Scanned 10/5/2026

$npx -y skills add polar-bear-org/claude-skills --skill gnet-reason-to-write --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Gnet Reason To Write?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Gnet Reason To Write
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/polar-bear-org-gnet-reason-to-write/badge)](https://www.skillsdirectory.com/skills/polar-bear-org-gnet-reason-to-write)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: gnet-reason-to-write
description: Picks one reason to write per person that is useful to them, why it fits, and what not to ask yet. Use for "run gnet-reason-to-write", "every message I start is an ask", "what do I say to them", "reason to message someone", "how to reach out without asking for a job", "I don't want to sound needy", "reconnect with an old contact", "why would they reply to me", part of the Claude Graduate Networking Guide Pack by Polar Bear.
---

# Reason to Write

## When To Use
You know who to contact, but every message you start is an ask: a job, a referral, a look at your CV. Run this before you write anything, to answer one question per person: what reason to write would be useful to them, and what should wait?

## When Not To Use
If you have not chosen who to contact yet, run Contact Shortlist first. If you already have a reason and a blank screen, go straight to Message Points; if the person is a stranger you have no way to message yet, Connection Request Note fits better.

## Inputs
- Your shortlist of about 10 people: name, role, employer, how you know them, and the closeness tier in your own words
- Anything real you share with each one: past help, a talk you heard, a course, a placement, a recent public post or piece of work, with its source
- Your Networking Brief, if you have one (the help you need, what you can offer back)
If you have none of this, I start from three names and how you know each one, and mark the output as a first draft.

## Approach
Reasons to write that help the other person, a principle described here in our own words, combined with the advice that runs through UK careers guidance and the TARGETjobs LinkedIn guide: ask for advice, not a job. The judgment is matching the reason to the relationship, so a former manager can take a direct ask for advice while an alumnus you have never met gets something lighter. The failure it prevents: the "congratulations on the new role" message to a stranger that turns into a job ask in line two, which both of you can see coming.

## Workflow
1. Ask at most three questions: which help from your brief matters most this month, whether anyone on the list has helped you before, and whether you have anything useful to share (an article, an event, a contact, notes from a lecture).
2. For each person, pick one reason from six: thank (for past help), share (something useful, no ask), congratulate (something recent, with a source), catch up (a dormant tie, no ask), ask for advice, or ask for a 15-minute conversation.
3. Test the reason against reality. It must rest on a real link or a sourced fact. A congratulation with no source, or a "shared interest" you guessed at, is rejected as a pretext and replaced or marked "no reason yet".
4. Match the weight of the reason to the tier. Close and warm ties can take a direct ask for advice or 15 minutes. Distant or never-met contacts get a thank, a share or a congratulation, or I suggest an introduction route through someone closer.
5. Fill the "not yet" column for every person: job, referral, CV review and internship asks wait. Never a job in the first message.
6. Show the table and let you approve, swap or drop each reason. Keep the run to about 10 people; the rest wait for the next run.

## Output Format
```markdown
# Reasons to Write
Run of [date] · [number] people · brief used: [yes or no]
## Reasons
| Person | How you know them | Reason (one of six) | Why it fits, with source | Not yet |
|---|---|---|---|---|
| [contact] | [link, in your words] | [thank / share / congratulate / catch up / advice / 15 minutes] | [the real link or sourced fact] | [job, referral, CV] |
## No reason yet
- [contact]: [what is missing, e.g. no shared link and nothing sourced] -> [find one, or ask someone closer to introduce you]
## Decision
[You approve, swap or drop each reason by [date], then pick the first person to write to.]
```

## Done When
- Every person has exactly one reason from the six, or sits under "No reason yet"
- Every reason traces to a real link or a fact with a source
- Every row has a "not yet" entry and no first message asks for a job or referral
- The run holds about 10 people, chosen by you

## Quality Bar
- One reason per person; two reasons in one message reads as a pretext for the second
- Reasons are about the relationship, never a judgement of who "deserves" contact
- Congratulations only for something public and recent, with its source
- Contact details stay out of the table and out of Claude memory
- Claude suggests reasons that help them; you pick one and nothing is invented as a pretext.

## Next
Run gnet-message-points (Message Points) to get points for the first message.

## About the makers

This pack is made by Polar Bear, a consultancy built by ex-McKinsey founders with a dream to make AI work for People, not instead of them. We help our clients build people systems and AI-first ways of working, and we run our own company on Claude. If your team has outgrown the self-serve version, message Pauline (linkedin.com/in/paulinebertry).

Attribution

polar-bear-orgpolar-bear-org
View sourceSee grades on GitHubMore from polar-bear-org →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698461 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →