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/tako-dev/cli/tako-agentnpx skills add tako-dev/cli --skill tako-agentgit clone --depth 1 https://github.com/tako-dev/cliWhat 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 | $0.00102 | $0.01383 |
| Opus 5 | $0.00051 | $0.00691 |
| Sonnet 5 | $0.00020 | $0.00277 |
| Haiku 4.5 | $0.00010 | $0.00138 |
Grade A, and why
tako-agent 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 yesterday.
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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
tako-agent — 长时 agent session 管理
tako agent <subcmd> [...]
快速速查
# 一次性任务（推荐脚本场景）：start+send+(可选)回收，一条命令
tako agent run <claude|codex> --prompt "..." [--model X] [--name N] [--cwd .] [--json] [--purge]
tako agent start <claude|codex> [--model X] [--name N] [--cwd .] [--approval yolo|external] [--json]
tako agent list [筛选] [--json] # 默认隐藏 closed；--all 显示全部
tako agent send <sid> "prompt..." # 阻塞发送一轮
tako agent show <sid> [--lines N] [--errors-only] # meta+日志；--errors-only 只看错误并展开详情
tako agent logs <sid> [--errors] [--json] # dump 完整日志（不截断），适合 grep/喂 LLM
tako agent attach <sid> # 实时 tail 日志流
tako agent cancel <sid> | cancel [筛选] # 中止 turn（带筛选则批量）
tako agent close <sid> [--purge] | close [筛选] [--purge] [--yes] # 批量关闭
tako agent purge # 清理 closed/dead
# 筛选 flag（用于 list / 批量 close|cancel）
# --status idle,running --name-prefix fg- --backend claude --model X --turns 0 --all --yes
# 外置审批（codex）
tako agent pending <sid> # 列待审批
tako agent approve <sid> <id> <allow|deny> [--reason "..."] [--rule "<regex>"]
tako agent wait <sid> [--json] # 阻塞到下一决策点（LLM 友好）
# 策略
tako agent policy <sid> show
tako agent policy <sid> allow-exec <regex>
tako agent policy <sid> deny-exec <regex>
tako agent default <claude|codex> <providerId>
典型工作流
派 agent 去做任务（一次性，推荐）
# run 一条命令搞定，--json 直接拿结果，无需 grep sid
RESULT=$(tako agent run codex --model gpt-5.5 --name research --json \
--prompt "扫描 src/ 所有 TODO，按文件分组列出")
echo "$RESULT" | jq -r '.text' # 取结果文本
echo "$RESULT" | jq -r '.status' # ok / error
# 默认保留 session 便于查日志；加 --purge 自动回收
多 turn 续接（需保持 session）
SID=$(tako agent start claude --model claude-opus-4-7 --name review --json | jq -r '.sid')
tako agent send "$SID" "review src/agent/manager.ts 的并发安全"
tako agent send "$SID" "给一个最小补丁" # 历史自动保留
并发多 session（批量管理）
# 用统一 name 前缀，便于事后批量筛选/清理
for grp in alpha beta gamma; do
tako agent run codex --name "fg-$grp" --json --prompt "处理 $grp" &
done
wait
tako agent list --name-prefix fg- # 只看这批
tako agent close --name-prefix fg- --status idle --yes # 批量收尾
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.
- yesterday First seen · 106 lines · 102 tokens per session scan A 97073b189d44
tako-agent is a skill published in the GitHub repository tako-dev/cli (3 stars, last pushed 2d ago), licensed MIT. It adds 102 tokens to every session and 1,383 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…