agentleFS
migavel508migavel508/wisprnote-skills/skills/commitment-tracker/SKILL.md

commitment-tracker

Skill0 starsChanged 16 days agoMITView on GitHub

Turn commitments made in conversation into tracked work in Jira, GitHub, Linear or a task list — one item at a time, with a human approving each before anything is created. Use after a meeting that produced action items, or when a backlog has drifted from what was actually agreed.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. legioncodeinc/littlebird-skills★ 8same name
    2. youdovhan/claude-skills-bundle★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository4 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Keep the original wording attached to each item.

    line 30
  2. Distinctive:

    Before proposing anything, search the tracker for a match on substance.

    line 34
  3. Distinctive:

    If something close exists, propose updating it rather than creating a new one.

    line 38
  4. Distinctive:

    If the user says "do them all", still show the full list once and get confirmation on that list before writing anything.

    line 67
  5. Distinctive:

    After each creation, report the identifier and URL.

    line 72
  6. Distinctive:

    Unowned items do not become tickets: If no person was named, surface it and ask who owns it.

    line 77
  7. Distinctive:

    If the tracker write fails, report the failure plainly and do not retry with altered content.

    line 85

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

Commitment tracker

The gap between "we agreed Alice would do X by Friday" and a ticket assigned to Alice due Friday is where most organisational follow-through dies. This skill closes it — without creating a pile of low-quality tickets nobody asked for.

The rule that matters

Never create, assign, or close anything without explicit approval for that specific item. Not batch approval. Not "yes, go ahead" once at the start. Each write is confirmed on its own.

Bad tickets are worse than no tickets: they create cleanup work, they make the tracker untrustworthy, and they are attributed to a person who never agreed to them. The expensive failure mode here is not missing one item — it is filing twelve wrong ones.

Procedure

1. Collect commitments

With Wisprnote: list_action_items, or get_meeting_knowledge for a specific meeting. Otherwise work from notes the user supplies.

Keep the original wording attached to each item. You will need it at the approval step.

2. Check whether it already exists

Before proposing anything, search the tracker for a match on substance. Duplicate tickets are the most common damage this kind of automation causes. Search the owner's name plus a key noun, not the full sentence.

If something close exists, propose updating it rather than creating a new one.

3. Draft, do not create

For each commitment, prepare a proposal and show it in full:

Proposed: <project> ticket

Title:       <imperative, specific — "Add retry to webhook handler", not "Webhooks">
Assignee:    <name> — as named in the meeting
Due:         <date> — as agreed, or "none agreed"
Description:
  <what needs doing, in enough detail to act on>

  Agreed in: <meeting>, <date>
  As stated: "<original wording, verbatim>"

Source confidence: high | medium — <whether owner and date were explicit>

The verbatim line is not decoration. It lets the approver catch the case where your interpretation drifted from what was actually said.

4. Ask, one at a time

Present each proposal and wait. Accept four outcomes: create it, edit then create, skip, stop entirely. Treat silence or ambiguity as skip.

If the user says "do them all", still show the full list once and get confirmation on that list before writing anything.

5. Create, then report back

After each creation, report the identifier and URL. At the end, summarise: what was created, what was skipped, what was already tracked.

Guardrails

  • Unowned items do not become tickets. If no person was named, surface it and ask who owns it. Assigning by inference puts work on someone who never agreed to it.
  • Never close or reassign existing work unless explicitly asked for that specific item. Creating is recoverable; closing someone else's ticket is not.
  • Respect "no date agreed". Inventing a due date manufactures a commitment nobody made.
  • Do not editorialise. The description carries what was agreed, not your assessment of whether it is a good idea.
  • If the tracker write fails, report the failure plainly and do not retry with altered content.

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.