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/tmdry4530/code-flow/flownpx skills add tmdry4530/code-flow --skill flowgit clone --depth 1 https://github.com/tmdry4530/code-flowWhat 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 | $0.00045 | $0.00525 |
| Opus 5 | $0.00023 | $0.00262 |
| Sonnet 5 | $0.00009 | $0.00105 |
| Haiku 4.5 | $0.00005 | $0.00052 |
Grade A, and why
flow 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
For each port from 4560 to 4570, request `http://localhost:<port>/api/health` (e.g. `curl -s -m 2`). If a response has `"app":"code-flow"` and its `projectRoot` equals the current project directory, the dashboard is alre How it starts
The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
code-flow dashboard launcher
Start (or reuse) the local code-flow dashboard server for the current project, then tell the user the URL. Follow these steps exactly.
1. Check for a running instance
For each port from 4560 to 4570, request http://localhost:<port>/api/health (e.g. curl -s -m 2). If a response has "app":"code-flow" and its projectRoot equals the current project directory, the dashboard is already running — report http://localhost:<port> to the user and STOP. Do not start a second server.
2. Server runtime (already prebuilt)
The plugin root is available as ${CLAUDE_PLUGIN_ROOT}. The dashboard ships prebuilt — a marketplace install clones this repo and runs no install/build. You should not need to build anything; requires only Node.js ≥ 20.
Expect these to already exist:
${CLAUDE_PLUGIN_ROOT}/server/dist/main.js— the bundled server (plusserver/dist/chunks/)${CLAUDE_PLUGIN_ROOT}/web/dist/index.html— the static UI
The Architecture and Data-flow views additionally need code-analysis packages (dependency-cruiser, ts-morph); the server installs those itself on first use, so you don't set them up here. The Timeline works without them.
Only if server/dist/main.js is missing (e.g. running from a source checkout that hasn't been packaged), build once from source — this is a developer fallback, not the normal path:
cd "${CLAUDE_PLUGIN_ROOT}" && pnpm install && pnpm bundle
3. Start the dashboard
Run in the background from the user's project directory:
CODE_FLOW_PROJECT="$(pwd)" node "${CLAUDE_PLUGIN_ROOT}/server/dist/main.js"
It prints either code-flow dashboard: http://localhost:<port> or, if another instance grabbed the port first, already running at <url>.
4. Report
Tell the user the dashboard URL and that new edits made by the agent will appear live (no refresh needed). If the project has no recorded sessions yet, mention that the dashboard fills up as soon as the agent edits files with the plugin enabled.
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 · 44 lines · 45 tokens per session scan A ab1c5f23e522
flow is a skill published in the GitHub repository tmdry4530/code-flow (1 stars, last pushed 1mo ago), licensed MIT. It adds 45 tokens to every session and 525 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
pre-landing-review
Pre-landing PR review. Analyzes diff against the base branch for SQL safety, LLM trust boundary violations, conditional side effects, and other structural issues. Use when explicitly asked for the specialized pre-landing workflow. Product /review requests are handled by BitFun's unified Review mechanism instead.…
wechat-article-writer
公众号/自媒体全流程。根据用户表述自动匹配：撰写文章、封面图、正文插图、风格提取。支持多种写作风格。当用户提到写公众号、技术博客、公众号封面、正文插图、步骤图、演示图、流程示意、分析写作风格、克隆文风、模仿爆款、提取风格时使用。详见 reference 目录。.
vibe-research
Deep research and market validation for app ideas. Use when starting a new project, validating an idea, or when the user says "research my idea", "validate my app", or "help me start a new project".
commit-push-pr
Commit selected local changes, push the branch, and create or update a GitHub pull request with BitFun attribution. Use when the user asks to 提交 PR、提代码、commit and push、开 PR、create a pull request, or wants a Claude Code-like one-command PR publishing flow from BitFun.
investigate
Systematic debugging with root cause investigation. Four phases: investigate, analyze, hypothesize, implement. Iron Law: no fixes without root cause. Use when asked to "debug this", "fix this bug", "why is this broken", "investigate this error", or "root cause analysis". Proactively invoke this skill (do NOT debug…
pr-review-canvas
Create a BitFun Canvas for reviewing a pull request, branch diff, or change set with Cursor-style diff cards, review maps, risk callouts, and focused reviewer flow. Use when the user asks for a PR review canvas, diff walkthrough, change-set overview, or visual review summary.