agent-task SKILL.md skillA
agent-task from zlin101/trellium is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add zlin101/trellium/agent-task@git:20260918.c284557Pins git:20260918.c284557 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- zlin101/trellium
- Path
docs/evals/adoption-durability-2026-09/materials/fixture-a/skills/agent-task/SKILL.md- Versions recorded
- 1
- Latest version
- git:20260918.c284557, committed Sep 18, 2026
- Size
- 3,528 bytes
- Repository stars
- 28
- License
- none declared
用于执行需要上下文读取、限定范围修改、验证、任务记录或 vault 记忆更新的非琐碎项目任务。.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Agent Task Workflow ## Steps 1. 读取 `AGENTS.md`、`vault/index.md`（含速查表）和 `vault/runtime.md`；Level B/C、判定模糊或涉及治理规则时读取 `vault/governance.md`。 2. 按 `vault/index.md` 读取任务特定上下文。 3. 将任务归类为 Level A、Level B 或 Level C。 4. 判断授权等级和是否需要用户确认。 5. 明确目标、范围、不做范围、验收标准和检查。 6. Level B 或 Level C 创建或更新任务文件。任务文件的 `trellium-task-state` 状态块是 lifecycle、authority_level、当前 slice 与 Gate 结果的唯一 owner：每次状态变化先更新状态块，再同步 `runtime.md` 对应行（投影）。没有状态块的旧任务文件是 legacy：下次接触时补上。 7. 做最小必要修改。 8. 行为变化时添加或更新聚焦测试。 9. 通过任务文件和 `vault/handoff.md` 让长任务可恢复。 10. 运行必要检查。 11. 检查验收门；测试通过不等于完成。 12. 多轮 review 使用 `vault/tasks/TASK-xxxx-review.md` 台账：findings 编号进入、批量处理、批量回写状态（open/fixed/wont-fix/needs-discussion）；收敛后归档进任务文件 Execution Record。 13. 更新 `vault/runtime.md`：只改 Active Tasks 表中对应任务行的状态与下一步，需要时调整 Focus 行；每条一行、单行替换，不重写整段。 14. 长期决策写入 `vault/decisions.md`。local 任务（`task_storage=local`）进入 `accepted` 前在任务文件 Memory Updates 填写 Durable knowledge disposition：`none — <理由>` 或 `distilled — <canonical 目标文件>`；未填写视为 `pending`，不得进入 `ready_for_review` 或 `accepted`。错误契约直接 `superseded`，不受该 gate 阻塞；local 任务关闭后删除 `runtime.md` 对应行。 15. 用户挂起任务或决定时，在 `vault/parked.md` 记条目（含重启触发器）；用户重新提起时升回任务文件（draft）或 `runtime.md`。 …
Read the whole file at its exact version.
How to install
mdr add zlin101/trellium/agent-task@sha256:bd927b6185ea52aeThe line at the top of this page pins the label git:20260918.c284557, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_hj4ijry5uojdozx2)
2 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Sep 18, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (3528 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
zlin101/trellium · 28 stars · license none declared · pushed 2026-10-01 · branch develop
API
GET https://markdownregistry.com/api/v1/artifacts/art_hj4ijry5uojdozx2 GET https://markdownregistry.com/api/v1/resolve?ref=zlin101/trellium/agent-task GET https://markdownregistry.com/api/v1/blob/bd927b6185ea52aef9164687fe58640d18cee21a7537d0a3fc41182c8d5188f1
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.
More from zlin101/trellium
Every file in zlin101/trellium