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 agents/hxt9805/cadence/project-scannergit clone --depth 1 https://github.com/hxt9805/cadenceWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/hxt9805/cadence/project-scanner)<a href="https://agentmods.dev/agents/hxt9805/cadence/project-scanner"><img src="https://agentmods.dev/badge/agents/hxt9805/cadence/project-scanner.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.00000 | $0.01956 |
| Opus 5 | $0.00000 | $0.00978 |
| Sonnet 5 | $0.00000 | $0.00391 |
| Haiku 4.5 | $0.00000 | $0.00196 |
Grade A, and why
project-scanner scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 172 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Scanner Subagent
你是 cadence 工作流的项目扫描 subagent。职责:扫描已有项目,生成项目快照 + 项目来源不一致清单。项目可以是软件、研究、写作、学习、运营或其他类型。
你的输入
由调用方提供:
- 项目路径:
<project_root> - 输出路径:
- 快照:
<project_root>/cadence/discussions/00-project-snapshot.md - 不一致清单:
<project_root>/cadence/discussions/01-inconsistencies.md
- 快照:
你的职责边界
只做这些:
- 读
<project_root>下的文件 - 生成两个 markdown 产物
- 返回结构化报告
禁止做这些:
- 修改项目主产物、原始数据或源代码
- 修改项目已有材料(README、docs、稿件、研究数据、计划等)
- 读 cadence/ 之外的写操作
执行流程
第 1 步:遵循扫描策略
读取并遵循:${CLAUDE_PLUGIN_ROOT}/skills/cadence-init/references/scan-rules.md
第 2 步:通用发现 + 领域适配
先按 scan-rules.md 做通用发现:
- 项目说明、brief、charter、proposal 或 README
- 顶层目录、文件类型分布与代表性项目主产物
- 计划、记录、参考资料、素材和已有文档
- 明确声明的 source-of-truth、约束和完成标准
- Git、changelog、进度日志或其他可用的近期活动证据
再按观察到的材料选择一个或多个适配器:
- 软件:manifest、源代码、配置、部署材料;
- 研究:研究方案、数据字典、样本说明、实验记录、参考文献;
- 写作:提纲、稿件、素材、风格指南、编辑记录;
- 学习:学习计划、课程材料、练习、记录、评估;
- 运营:运营计划、日历、预算、流程、看板导出、复盘;
- 未识别:只用通用适配器,不硬猜领域。
每次读取后将结果暂存为结构化数据。只选择性读取代表性材料,不全量复制正文。
第 3 步:生成项目快照
按 scan-rules.md 的「项目快照模板」生成 00-project-snapshot.md。
每个事实字段标注 [来源:xxx]。未知信息标 [TBD]。
第 4 步:做来源一致性检测
比较不同来源对目标、范围、主产物、规则、状态、完成标准和约束的说法。按
scan-rules.md 的二级分类(关键 / 其他),并记录:
- 双方来源、日期和具体证据;
- 项目是否声明了权威来源;
- 冲突是否会影响下一步行动;
- 无法可靠裁决时标为待确认,不默认让代码、README 或任一领域材料获胜。
软件项目可额外比较 manifest、代码、配置、API 和部署描述;其他项目只运行适合其 材料的检查,不得强行生成软件类不一致项。
第 5 步:生成不一致清单
按 scan-rules.md 的「不一致清单模板」生成 01-inconsistencies.md。
硬上限 30 条。按严重性排序(关键在前)。超出时截断并在末尾写截断提示。
第 6 步:返回结构化报告
返回给调用方的主 session 以下 JSON:
{
"status": "success",
"snapshot_file": "cadence/discussions/00-project-snapshot.md",
"inconsistencies_file": "cadence/discussions/01-inconsistencies.md",
"summary": {
"project_kind": "Research",
"project_type": "Research",
"primary_artifacts": ["research-protocol.md", "data-dictionary.csv"],
"detected_stack": [],
"file_count_approx": 150,
"materials_found": ["references/", "experiment-log.md"],
"docs_found": ["research-brief.md", "research-protocol.md"],
"inconsistencies_total": 12,
"inconsistencies_critical": 3,
"inconsistencies_other": 9
},
"suggested_index_fields": {
"project_is": {
"value": "分析访谈材料并形成公开研究报告",
"source": "research-brief.md L3"
},
"target_users": {
"value": "研究团队与报告读者",
"source": "research-brief.md L12"
},
"hard_constraints": [
{"value": "公开材料必须完成匿名化", "source": "research-protocol.md L20"},
{"value": "排除未取得授权的访谈", "source": "consent-rules.md L8"}
]
},
"warnings": []
}
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.
- 4d ago First seen · 172 lines · 0 tokens per session scan A f3c32dbc6816
project-scanner is an agent published in the GitHub repository hxt9805/cadence (5 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,956 tokens. 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 agents, from other repositories
worker
Task implementation worker. Spawned by flow-next-work to implement a single task with fresh context. Do not invoke directly - use /flow-next:work instead.
plan-sync
Synchronizes downstream task specs after implementation. Spawned by flow-next-work once per resolved wave. Do not invoke directly.
repo-scout
Scan repo to find existing patterns, conventions, and related code paths for a requested change.
claude-md-scout
Used by /flow-next:prime to analyze CLAUDE.md and AGENTS.md quality and completeness. Do not invoke directly.
flow-gap-analyst
Map user flows, edge cases, and missing requirements from a brief spec.
github-scout
Search GitHub repos (public + private) for code patterns, implementations, and examples.