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/aisee-lab/aisee-plugin/aisee-knowledge-curatenpx skills add AISEE-LAB/aisee-plugin --skill aisee-knowledge-curategit clone --depth 1 https://github.com/AISEE-LAB/aisee-pluginWrote 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/skills/aisee-lab/aisee-plugin/aisee-knowledge-curate)<a href="https://agentmods.dev/skills/aisee-lab/aisee-plugin/aisee-knowledge-curate"><img src="https://agentmods.dev/badge/skills/aisee-lab/aisee-plugin/aisee-knowledge-curate.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 5.1 | $0.00116 | $0.01034 |
| Opus 5 | $0.00058 | $0.00517 |
| Sonnet 5 | $0.00023 | $0.00207 |
| Haiku 4.5 | $0.00012 | $0.00103 |
Grade A, and why
aisee:knowledge-curate 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 6d 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 — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
aisee:knowledge-curate
批量审查项目内候选知识,产出可人工 review 的 team knowledge card drafts。
职责
- 读取项目内
aisee/docs/reflect/knowledge-candidates/。 - 按需读取相关
docs/solutions/frontmatter 或摘要证据。 - 去除项目私密信息、客户信息、内部 URL、secrets 和不可公开路径。
- 泛化候选,使其适用于相同技术栈、schema、phase、surface 或 risk 类型。
- 合并重复候选,标记 stale candidate。
- 输出 batch review report 和 card drafts。
不负责
- 不替代
aisee:reflect;reflect 负责生成项目内候选。 - 不替代
ce-compound;具体工程问题 solution 仍由 Compound 记录。 - 不复制 solution 正文。
- 不把 candidate 自动提升为 active card。
- 不自动写入
aisee-team-knowledge;用户授权后可以建议运行aisee knowledge promote-batch。 - 不手工创建 team knowledge 仓库骨架或直接改业务项目
aisee/knowledge.yaml;初始化和接入优先走aisee knowledge init-repo与aisee knowledge configure。 - 不创建分支、commit、push、merge 或 PR,除非用户明确授权并提供 team repo 路径。
- 不让 team knowledge 覆盖 OpenSpec specs、tasks、contracts、source-map 或 baseline。
Phase 0 — 扫描候选
扫描时遵守 .gitignore,优先使用 rg --files:
rg --files aisee/docs/reflect docs/solutions 2>/dev/null | rg 'knowledge-candidates|solutions|reflect'
如果没有 project-local candidates,只输出“没有可审查候选”,不要临时从聊天记录或全仓库生成 team card。
Workflow
按需读取 references:
- 执行批量审查时读取
references/workflow.md。 - 生成正式 batch review report 时读取
references/batch-review-template.md。 - 需要确认 card 字段时读取
references/knowledge-card-contract.md或项目根目录references/knowledge-card-contract.md。
默认流程:
- 收集候选文件和 evidence 引用。
- 过滤一次性、项目专有或证据不足的候选。
- 对剩余候选做去敏、泛化和边界补全。
- 与已配置 team knowledge active cards 做概念去重;没有配置时只做项目内去重。
- 生成 batch review report。
- 仅当用户明确要求写文件时,写入
aisee/docs/reflect/knowledge-curation/YYYY-MM-DD_<slug>.md。
Guardrails
- 默认只输出审查报告,不改 team repo。
- 写入 team repo 前必须再次确认用户授权和目标路径;优先使用
aisee knowledge promote-batch --curation <path> --team-path <path> --pack <id> --json写入本地 worktree。 - 如果用户还没有 team knowledge 仓库,先建议
aisee knowledge init-repo --dest <path> --initial-pack <id> --json;如果业务项目还没 pin,建议aisee knowledge configure --path <path> --enable-pack <id> --json。 promote-batch不会创建分支、commit、push、merge 或 PR;这些 Git 动作仍需用户明确授权。- 推荐 batch review:积累 3-10 条真实可复用候选后再提交。
- 安全、高风险或公开接口类候选可以建议单独 PR,但仍需用户授权。
- 每条 draft 必须包含 required machine fields:
id、title、status、applies_to、trigger、recommended_action、boundaries。 evidence、risk_types、tags是 review 信息,不是 active card 硬必填。- 最终回复列出候选数量、合并数量、拒绝数量、draft 数量、敏感信息风险和写入路径。
What ships with it
15 files 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.
- agents/openai.yaml 303 B
- assets/team-knowledge/.aisee-team-knowledge 20 B
- assets/team-knowledge/AGENTS.md 313 B
- assets/team-knowledge/docs/authoring-guide.md 260 B
- assets/team-knowledge/docs/review-policy.md 281 B
- assets/team-knowledge/knowledge/cards/cli/cli-json-output-stability.md 752 B
- assets/team-knowledge/knowledge/cards/openspec/openspec-source-map-is-routing.md 769 B
- assets/team-knowledge/knowledge/packs/openspec.yaml 228 B
- assets/team-knowledge/knowledge/packs/web-app.yaml 208 B
- assets/team-knowledge/README.md 341 B
- assets/team-knowledge/schemas/knowledge-card.schema.json 537 B
- assets/team-knowledge/schemas/knowledge-pack.schema.json 326 B
- evals/evals.json 1.6 KB
- references/batch-review-template.md 1.5 KB
- references/workflow.md 2.6 KB
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.
- 6d ago First seen · 68 lines · 116 tokens per session scan A eba9fa329749
aisee:knowledge-curate is a skill published in the GitHub repository AISEE-LAB/aisee-plugin (11 stars, last pushed 2mo ago), licensed MIT. It adds 116 tokens to every session and 1,034 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
codex-sdd:archive
Preserve change artifacts (proposal, design, tasks) into structured archive and update knowledge base. Final phase of OpenSpec workflow.
obsidian-vault
Organize Obsidian vaults with MOCs, wikilinks, frontmatter/properties, dashboards, orphan-note checks, and cleanup workflows. Use ONLY when the user explicitly refers to Obsidian — by naming "Obsidian", an Obsidian "vault", or an Obsidian-specific feature such as wikilinks ("[[ ]]"), the Dataview or Bases plugins, or…
task-agent-eliza-bridge
Use when spawning a Claude Code, Codex, Gemini, Aider, or other CLI task agent whose work needs parent Eliza runtime context. Covers the read-only loopback bridge for character, room, memory, and active workspace state.
aatmf-t04-memory-manipulation
AATMF T4 — Multi-Turn & Memory Manipulation. Persistent memory injection, conversation-state poisoning, cross-session contamination, ghost-context leak.
hive.note-taking
Maintain a free-form scratchpad of decisions, extracted values, and open questions so context pruning doesn't lose anything you still need.
memory
Use when the user asks to remember, recall, forget, update, search, or inspect durable OpenSquilla memory, including profile facts in USER.md and long-term notes in MEMORY.md or memory//.md.