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/tenkenco/declaude/setupnpx skills add tenkenco/declaude --skill setupgit clone --depth 1 https://github.com/tenkenco/declaudeWhat 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 | $0.00049 | $0.01104 |
| Opus 5 | $0.00024 | $0.00552 |
| Sonnet 5 | $0.00010 | $0.00221 |
| Haiku 4.5 | $0.00005 | $0.00110 |
Grade A, and why
setup scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s -w '\nHTTP %{http_code}\n' https://speak-english.tenken.co/v1/usage \ How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
declaude setup
The plugin already registers the Claude Code hook. It ships hooks/hooks.json, which
Claude Code loads on install. An unset hook_enabled option means on, so a fresh
install rewrites replies after its API key is configured. The manifest does not declare
a default because Claude Code may export declared defaults to plugin processes. The
api_key option stays optional because the MCP tools sign in on their own.
One case still stays inert. A user who exports DECLAUDE_TOKEN and never sets
hook_enabled keeps the plugin hook off. That user may still run a version 1.0 manual
hook, and two hooks would translate and bill every reply twice.
Work through the steps below in order. Report the result of each one.
1. Remove any manual hook entry
Claude Code merges hooks from several settings files. Search all of them:
for f in "${CLAUDE_CONFIG_DIR:-$HOME/.claude}"/settings.json \
"${CLAUDE_CONFIG_DIR:-$HOME/.claude}"/settings.local.json \
.claude/settings.json .claude/settings.local.json; do
[ -f "$f" ] && grep -l declaude_hook "$f"
done
Open each file the search names. Look for a MessageDisplay or Stop entry that runs
declaude_hook.py. Remove that entry.
The plugin now registers the hook. A manual entry on top of it translates every reply twice and bills the user twice. Do not enable the plugin hook until every manual entry is removed.
2. Check for a legacy key
echo "${DECLAUDE_TOKEN:+set}"
If the answer is set, do not edit or replace it. The plugin can keep using
DECLAUDE_TOKEN during migration. If the answer is empty, the user may still have a key in
secure plugin storage; check that in step 3.
3. Configure and enable the plugin hook
Only after every manual entry is removed, run:
/plugin configure declaude@tenken
Set the options in the configuration dialog:
api_key: if a key is already stored, do not edit or replace it. Otherwise get adk_key at /signin and enter it in this masked field. Do not paste it into the chat. ExistingDECLAUDE_TOKENusers may leave this empty, because the hook falls back to that environment variable.hook_enabled: typetrueorfalsein this field. An unset value means on. Typefalseto stop automatic rewrites. Typetrueif the user exportsDECLAUDE_TOKENand wants the plugin hook on. The hook exits silently when it has no key, so a missing key produces no error message.
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 · 102 lines · 49 tokens per session scan A 21a402a944be
setup is a skill published in the GitHub repository tenkenco/declaude (1 stars, last pushed 3d ago), licensed MIT. It adds 49 tokens to every session and 1,104 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
story
网络小说工具箱主入口。根据用户需求自动路由到对应 skill,并可管理作者习惯、启动本地 Dashboard。触发方式:/story、$story、/story dashboard、/网文、「我想写小说」「记住我的写作习惯」「打开工作台」「检查更新」。.
datamodel-code-generator
Use this skill when the user wants Python data models, Pydantic models, dataclasses, TypedDicts, msgspec structs, or type-safe Python classes generated from OpenAPI, AsyncAPI, JSON Schema, GraphQL, JSON/YAML/CSV sample data, MCP tool schemas, Protocol Buffers, XML Schema, Apache Avro, or existing Python model objects.…
openai-docs
Use when the user asks how to build with OpenAI products or APIs and needs up-to-date official documentation with citations (for example: Codex, Responses API, Chat Completions, Apps SDK, Agents SDK, Realtime, model capabilities or limits); prioritize OpenAI docs MCP tools and restrict any fallback browsing to…
aws-wechat-article-topics
公众号选题|爆款标题|热点追踪|系列策划 — 公众号 AI 选题与标题生成,覆盖热点调研、选题策划、起标题、写摘要、系列排期。面向自媒体编辑、内容运营。触发词(单独触发仅限对已有标题/摘要的修改):「改标题」「换个标题」「重起标题」「优化标题」「标题再想想」「换个标题试试」「改摘要」「重写摘要」「优化摘要」「摘要再优化下」。新做选题、起新标题、策划系列/内容日历、追热点都请走 aws-wechat-article-main;需要多环节串联(写+审+排+配图+发)也走 main。.
style-modeler
只根据用户提供或明确允许获取的样本建模。把作者身份用于样本分组和风格库检索,不把外部履历、既有印象或样本之外的观点写进风格档案。.
trellis-start
Initializes an AI development session by reading workflow guides, developer identity, git status, active tasks, and project guidelines from .trellis/. Classifies incoming tasks and routes to brainstorm, direct edit, or task workflow. Use when beginning a new coding session, resuming work, starting a new task, or…