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 agentmods add skills/langchain-ai/deepagents/coding-prefsnpx skills add langchain-ai/deepagents --skill coding-prefsgit clone --depth 1 https://github.com/langchain-ai/deepagentsWhat 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 | $0.00036 | $0.00281 |
| Opus 5 | $0.00018 | $0.00140 |
| Sonnet 5 | $0.00007 | $0.00056 |
| Haiku 4.5 | $0.00004 | $0.00028 |
Grade A, and why
coding-prefs 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.
What it actually says
Coding Preferences Skill
Use this skill to keep /memory/coding-prefs.md in sync with how this
specific user wants you to work. This file is user-scoped, so each user
has their own copy — anything you write here only affects future
conversations with the same user.
When to Read
- Before picking a code style, test framework, or commit message format
- Before deciding whether to add comments, type hints, or docstrings
- Before refactoring beyond what was asked
When to Write
Append a new entry whenever the user gives feedback that should apply to future work:
- "Don't add docstrings unless I ask" → save it
- "I prefer pytest over unittest" → save it
- "Stop summarizing what you did at the end" → save it
Each entry should be one line: the rule, then a brief reason if the user gave one.
How to Write
Read the file first (it may not exist yet), then append. Don't overwrite — preferences accumulate over time. If a new preference contradicts an existing one, replace the old line and note the change.
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 · 36 lines · 36 tokens per session scan A bd25243f2f61
coding-prefs is a skill published in the GitHub repository langchain-ai/deepagents (28,825 stars, last pushed today), licensed MIT. It adds 36 tokens to every session and 281 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
code-reviewer
当用户要求进行代码审查、Code Review、查找 Bug、安全风险、性能问题或代码质量问题时使用。.
emoji-translator
当用户明确要求把自然语言翻译成 Emoji、把 Emoji 解释成文字，或要求“表情翻译/emoji 翻译”时使用。.
defense-evasion
Endpoint defense bypass — AMSI/ETW patching, ScareCrow framework, custom loaders, direct/indirect syscalls, LOLBAS execution, process injection.
opsec
Operational security management — traffic shaping, scan rate limiting, source IP management, tool signature avoidance, evidence handling, anti-detection patterns.
benchmark
Benchmark mode marker — engagement objective is flag capture. Generic engagement rules apply unchanged.
seven-question-gate
7-question gate run before promoting a finding to FINDING + opening a report. Kills weak/non-impactful findings before they reach the report stage and damage validity ratio.