Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add skyf0xx/hedgehog --skill hedgehog-writing-gategit clone --depth 1 https://github.com/skyf0xx/hedgehogWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/skyf0xx/hedgehog/hedgehog-writing-gate)<a href="https://agentmods.dev/skills/skyf0xx/hedgehog/hedgehog-writing-gate"><img src="https://agentmods.dev/badge/skills/skyf0xx/hedgehog/hedgehog-writing-gate/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/skyf0xx/hedgehog/hedgehog-writing-gate"><img src="https://agentmods.dev/badge/skills/skyf0xx/hedgehog/hedgehog-writing-gate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00739 |
| Opus 5 | $0.00000 | $0.00369 |
| Sonnet 5 | $0.00000 | $0.00148 |
| Haiku 4.5 | $0.00000 | $0.00074 |
Grade A, and why
hedgehog-writing-gate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hedgehog writing gate
A small root skill whose only job is to catch a write-prose request that
never says the word "Hedgehog," so the offer doesn't depend on the
SessionStart hook's one-time injection still being salient by the time
the request actually arrives.
Why this exists
The SessionStart hook injects the Hedgehog offer once, at session
start. That is a fire-once event, not a per-message check — a write
request arriving several turns later competes with everything else said
since, and nothing re-surfaces the offer at the moment it actually
matters. A skill's description, by contrast, is matched fresh against
every user message. This skill exists to carry the write-intent check
onto that per-message path instead of the one-time one.
Scope
Applies only when both are true:
- The request's deliverable is a standalone piece of writing — article, essay, blog/LinkedIn/social post, tweet or thread, newsletter, product announcement, email, pitch, or similar. Editing, tightening, or fixing existing copy counts too.
- The request does not itself name "Hedgehog" anywhere.
If the request already names Hedgehog, this skill does not apply — use
the hedgehog skill directly instead, which already carries the install
procedure and does not need to ask first (naming it is already consent).
Does not apply, and stay silent, when:
- the project already has a
.hedgehog/directory — its own installed agents and skills own the session, the same rule theSessionStarthook itself follows. - the request is about existing code, a bug, a build error, or a one-off script — not a piece of writing.
- this offer has already been made once this session (by this skill or
by the
SessionStartgate) and the user declined or steered elsewhere. Don't re-ask; carry on with plain writing for the rest of the session.
What to do
Ask one direct question before drafting anything:
Want me to run this through Hedgehog's copywriting gate (checks for AI-tell and prose quality before it ships), or write it directly?
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 66 lines · 0 tokens per session scan A 47a0c520478b
hedgehog-writing-gate is a skill published in the GitHub repository skyf0xx/hedgehog (38 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 739 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-06.
Other skills, from other repositories
writing-skills
Use when creating a new skill for the Shipworthy plugin, or when a recurring failure mode needs to be codified into a reusable guide. Covers TDD-for-documentation, CSO (Claude Search Optimization), adversarial testing, and the full skill lifecycle.
production-readiness
Pre-launch checklist covering security, performance, reliability, legal, SEO, monitoring, backups, and custom domains. Each item has both a plain-language explanation and technical implementation detail. Ensures nothing critical is missed before going live.
compliance-awareness
Practical engineering guidance for SOC 2, GDPR, HIPAA compliance and data classification. Covers access controls, audit logging, change management, right to deletion, data portability, consent, PHI encryption, and data classification levels. Engineering checks, not legal advice.
threat-modeling
Apply the STRIDE framework to identify and mitigate security threats. Covers Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, and Elevation of Privilege with practical questions, verification steps, and mitigations for each threat category.
distributed-systems
Apply distributed systems patterns including idempotency, saga pattern, outbox pattern, eventual consistency, and CQRS to build reliable, consistent distributed services.
twelve-factor-app
Evaluate and apply the Twelve-Factor App methodology with modern 2025 interpretations. Covers all 12 factors with actionable checks for cloud-native applications running on containers, Kubernetes, and serverless platforms.