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/keefar/ticket-flow/discovernpx skills add keefar/ticket-flow --skill discovergit clone --depth 1 https://github.com/keefar/ticket-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 | $0.00062 | $0.01419 |
| Opus 5 | $0.00031 | $0.00709 |
| Sonnet 5 | $0.00012 | $0.00284 |
| Haiku 4.5 | $0.00006 | $0.00142 |
Grade A, and why
discover 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 3d 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ticket-flow:discover — Project-Convention Discovery
Args: none — operates on current working directory.
What it does
Codifies the answer to "what conventions does this project follow?" into a single file that the harness itself loads. Claude Code discovers every .md under .claude/rules/ recursively and loads a rule without a paths: frontmatter key at launch, at the same priority as .claude/CLAUDE.md (memory docs) — so the conventions are in context for /ticket-flow:spec, /ticket-flow:implement and any plain session, without a skill having to fetch them. (Earlier versions wrote docs/PROJECT-CONVENTIONS.md, which only helped whoever remembered to read it.)
Adapted from Weselow Claude-Protocol's /project-discovery skill, but stripped of the rule-injection mechanism — output is a doc, not a hook.
Output
.claude/rules/project-conventions.md — sections below. An existing file is not overwritten: the script reports it and exits 1 — re-run with --force to replace it (there is no interactive prompt, so an agent must not read that exit code as a failure). A leftover docs/PROJECT-CONVENTIONS.md from an earlier run is reported and left alone — no skill reads it any more; merge what you want to keep and delete it.
Sections in the rule file
- Tech Stack — frameworks/languages/versions detected from manifest files
- Sources:
package.json,Cargo.toml,pyproject.toml/requirements.txt,go.mod,Package.swift,pom.xml,Gemfile,composer.json,*.csproj - Output:
Language: Swift 6.x · Framework: SwiftUI + AppKit · Min target: macOS 14+
- Sources:
- Build/Test/Lint commands — extracted from the manifests + common scripts (
Makefile,package.jsonscripts,xcodebuildinvocations in CI configs) - Naming conventions — sample 20 files, derive patterns
- SwiftUI:
*View.swift,*Model.swift, etc. - JS/TS:
kebab-casevscamelCasefor files - Python: PEP 8 module names
- SwiftUI:
- Testing setup — test framework + folder convention
Tests/+ XCTest +*Tests.swift|*Tests.m(Apple)tests/+ pytest +test_*.py(Python)__tests__/+ Jest (JS)
- Anti-patterns specific to this project — what not to do (extracted from existing CLAUDE.md, README, comments saying "// DO NOT")
What ships with it
1 file 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.
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.
- 3d ago First seen · 79 lines · 62 tokens per session scan A 27ba80e887dd
discover is a skill published in the GitHub repository keefar/ticket-flow (4 stars, last pushed 7d ago), licensed MIT. It adds 62 tokens to every session and 1,419 once invoked, about $0.0003 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
project-discuss
⚠️ 本 skill 与 brainstorming / writing-plans / executing-plans 等流程 skill 正交、可并行触发，不互斥。 已调用流程 skill 不等于可以跳过本 skill；只要涉及项目讨论/决策/查询，都需要额外激活本 skill。 项目讨论管理协议。管理讨论上下文、决策记录、查询行为。 触发：用户讨论项目方向/设计/技术选型/需求分析/功能规划时； 用户问"我们讨论到哪了"/"XX 确定了吗"； 用户想修改已确定的内容；用户提供外部文档； 用户询问项目现状/架构/实现细节/历史决策（含 debug 揭示架构问题时）； 当 bug…
cadence-bootstrap
Use when starting any conversation in a cadence-managed project (project root contains cadence/INDEX.md or cadence/ACTIVE.md). Establishes cadence workflow conventions — 记 / 整 / 查 三阶段记录协议, ACTIVE.md / INDEX.md state contract, domain-neutral adaptive recording fidelity, session start behaviors, and subagent dispatch…
cadence-resume
继续之前某次 session 的讨论上下文(v0.4:Step 6 archive cleanup B1 修复 + contenthashes 判据 + 派 retriever 替代 auditor)。用户说 '继续上次'/'resume' 或跑 /cadence-resume 时触发。.
cadence-handoff
整理当前 session 讨论内容到 cadence 档案(v0.4:触发整 阶段（ε 整合）兜底 + 写 sha1 书签式 snapshot,15-30 行)。用户说 'handoff'/'整理 session'/'换新 session' 或跑 /cadence-handoff 时触发。.
cadence-init
初始化 cadence 工作流目录结构。支持两种模式：新项目（极简模板 + TBD 占位） 和已有项目（通用发现 + 领域适配扫描 + 来源不一致清单）。适用于软件、研究、 写作、学习、运营及其他项目。当用户说"初始化 cadence"、 "开始用 cadence"、或跑 /cadence-init slash command 时触发。.
polish
Polish a completed code change with a bounded independent review protocol. Use after implementation and before final approval or PR to review intent, acceptance criteria, non-goals, diff, relevant files, verification, and previously rejected findings.