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 jzfgo/agents --skill write-like-megit clone --depth 1 https://github.com/jzfgo/agentsWrote 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/jzfgo/agents/write-like-me)<a href="https://agentmods.dev/skills/jzfgo/agents/write-like-me"><img src="https://agentmods.dev/badge/skills/jzfgo/agents/write-like-me.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00202 | $0.01583 |
| Opus 5 | $0.00101 | $0.00792 |
| Sonnet 5 | $0.00040 | $0.00317 |
| Haiku 4.5 | $0.00020 | $0.00158 |
Grade A, and why
write-like-me 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 2d 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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Write like me
This skill is generic and identical for everyone. Everything author-specific lives in a profile outside it, so the skill can be updated without touching the profile and the profile never has to be published.
Resolve the profile first
Before anything else, look for the profile, in this order:
.write-like-me/in the current project — a project can pin its own voice.~/.write-like-me/— the user's default.
The local one wins outright when both exist; do not merge them.
If neither exists, stop and say so. Do not improvise a voice, and do not start an extraction on your own:
There's no voice profile yet, so I can't write in your voice. Run
/write-like-me initand I'll build one from a corpus of your own writing — it's an interactive session and takes a while.
That is the whole response. init is long and expensive, and starting one the
user didn't ask for is worse than answering nothing.
Choose a mode
An explicit mode word wins. Otherwise infer it.
init Build the profile from a corpus. Only on the literal word.
update Correct one rule in an existing profile. Cheap; prefer it.
rewrite Rewrite an existing text in their voice.
edit Touch the minimum; keep their structure and order.
review Say what doesn't sound like them, and why. Do NOT touch the text.
(no mode) Write from scratch in their voice.
Ask only when review and rewrite are equally plausible, because handing a
rewritten text to someone who asked for a critique is the most expensive of the
four mistakes: they lose the decision.
init fires on the word, never on inference — update is what you almost
always want. A complaint about the voice ("it keeps using dashes", "that
doesn't sound like me any more") is an update: one labelled fix to one section
of VOICE.md, plus a regression case, in five minutes. A full extraction is
23 minutes and a large token bill, and it discards every correction the author
has made since. Only re-extract when the corpus itself was wrong.
What ships with it
12 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- .claude-plugin/plugin.json 441 B
- assets/en/llm-isms.md 4.6 KB
- assets/es/llm-isms.md 9.2 KB
- assets/profile-template.md 5.5 KB
- assets/shared/structural-tells.md 4.9 KB
- evals/evals.json 17 KB
- evals/trigger_eval.json 5.3 KB
- references/corpus-qualification.md 9.9 KB
- references/dimensions.md 7.3 KB
- references/init.md 23 KB
- references/regression-suite.md 4.4 KB
- references/update.md 3.0 KB
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.
- 2d ago First seen · 146 lines · 202 tokens per session scan A cd94e4da32e9
write-like-me is a skill published in the GitHub repository jzfgo/agents (1 stars, last pushed 4d ago), licensed MIT. It adds 202 tokens to every session and 1,583 once invoked, about $0.0010 per session on Opus 5. 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 2026-09-05.
Other skills, from other repositories
conductor-implement
Executes the tasks defined in the specified track's plan. Use this to start or continue working on a feature, bug fix, or chore.
ops-demo
CocoOps demo mode activator — populates .cocoplus/ops/demo/ with realistic mock data and sets cocoplus.toml [demo] enabled = true. Invoked via $ops demo.
agent-teams-simplify-and-harden
Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or…
acl-rule-analysis
Vendor-agnostic ACL and firewall rule analysis with shadowed rule detection, overly permissive rule identification, unused rule discovery, redundant rule flagging, and rule ordering optimization. Covers ACLs (Cisco/JunOS/EOS) and firewall policies (PAN-OS/FortiGate/CheckPoint).
add-analytics
Add Google Analytics 4 tracking to any project. Detects framework, adds tracking code, sets up events, and configures privacy settings.
agent-team-orchestration
Orchestrate multi-agent teams with defined roles, task lifecycles, handoff protocols, and review workflows. Use when: (1) Setting up a team of 2+ agents with different specializations, (2) Defining task routing and lifecycle (inbox → spec → build → review → done), (3) Creating handoff protocols between agents, (4)…