agentleFS
zhenzoozhenzoo/link16-agent-infra/.agents/skills/feishu-workline/SKILL.md

feishu-workline

Skill4 starsChanged 4 days agoApache-2.0View on GitHub

Link16 explicitly invokes this skill at the start of every Feishu turn to choose and commit the semantic card title and progress handoff.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before any other tool call or substantive work, read the current request and choose exactly one action:

    line 11
  2. Very specific:

    Run the exact session_work.py decide command supplied by the injected contract.

    line 21
  3. Distinctive:

    Do not narrate this protocol to the user.

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Feishu Workline

This is an explicit Link16 bridge protocol skill. Use it only when the bridge injects a [Link16 feishu-workline] turn contract.

Before any other tool call or substantive work, read the current request and choose exactly one action:

  • keep: the request continues the same concrete task and the existing semantic workline is still accurate.
  • replace: this is a new task, or the existing project/task is missing or no longer accurate. Supply project, task, and progress.
  • progress: the concrete task is unchanged but its visible Stage chain has materially changed. Supply the new progress.

Run the exact session_work.py decide command supplied by the injected contract. The semantic fields are model-owned:

  • project: the short project or repository label the owner recognizes.
  • task: a concrete object + action + deliverable. A reader must understand the work without opening the chat. Do not write placeholders such as “implementation and tests”.
  • progress: the current Stage chain with ✅ / 🔄 / ⏳; when a structured plan exists, preserve its meaning and order.

The bridge owns identity, session, runtime, route, timestamps, turn key, revision, status, card layout, colors, truncation, and summary. Never invent or override those fields. A keep decision is valid only when an earlier model-written workline exists. If the command rejects a stale turn or missing field, correct the semantic fields and retry once before continuing.

Do not narrate this protocol to the user. After the command succeeds, continue the original request normally.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.