agentleFS
Sign inStart free
h3y6eh3y6e/agent-skills/skills/tracking-tasks/SKILL.md

tracking-tasks

Skill0 starsChanged 28 days agoMITView on GitHub

Tracks task state and findings as Obsidian task/daily notes so work survives across sessions. Use when starting any multi-step task, resuming prior work or referring to an earlier task/issue/PR, switching direction or scope mid-task, wrapping up or reporting task status, or hitting a TIL, workaround, env/tooling quirk, or cross-cutting decision worth recording — even in a single-response task with no note yet.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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:

    When resuming a task, inspect linked notes/backlinks before broad search.

    line 33
  2. Distinctive:

    When logging a cross-cutting finding in the daily note, link the active task and relevant project/source pages there too.

    line 34
  3. Distinctive:

    If a finding is both task-relevant and cross-cutting, log task impact in the task note and the broader concern in the daily note separately.

    line 44

≈ “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

Tracking Tasks

Vault

~/ghq/github.com/h3y6e/memex — if missing, ghq get memex.

Task Note

Path: task/YYYY-MM-DD-<slug>.md. Frontmatter: title, status (backlog→todo→in-progress→done/canceled). Sections: Goal, DoD (checklist), Research, Notes.

  • Infer slug and title from context — don't ask.
  • On session start: search with anna recall "<text>", then read the matching file directly.
  • Fill Goal and DoD before substantial work, then set status to in-progress.
  • The note is the source of truth. Sync after initial plan, each work batch, direction changes, and before final response.
  • Evidence before claims — don't mark done until verification is recorded.

Graph

Make the Obsidian graph useful with [[wikilinks]]; do not treat notes as isolated logs.

  • Link concrete context: active project/repo, related task notes, daily notes, issues/PRs, and ADRs/specs.
  • When resuming a task, inspect linked notes/backlinks before broad search; prefer existing pages over new abstract concept nodes.
  • When logging a cross-cutting finding in the daily note, link the active task and relevant project/source pages there too.

Daily Note

Triggers are session-wide — even without a task note, even on single-response tasks. Write at the moment of discovery, not batched at the end.

Path: obsidian daily → daily/YYYY-MM-DD.md. If CLI fails, use ~/ghq/github.com/h3y6e/memex/daily/YYYY-MM-DD.md.

Triggers — TIL, workarounds, cross-cutting decisions, env/tooling issues, useful links, recurring patterns.

If a finding is both task-relevant and cross-cutting, log task impact in the task note and the broader concern in the daily note separately.

Obsidian CLI

The CLI is unstable. If any command errors, immediately fall back to direct file editing — do not retry or debug. Prefer direct editing over obsidian append for backticks or shell-sensitive content.

Workflow

The sections above say what to write and when; these are the exact commands.

  1. obsidian daily → today's date (use this, not the date command)
  2. anna recall "<query>" for an existing task; if none, obsidian create path="task/YYYY-MM-DD-slug.md" template=task
  3. obsidian property:set path="task/..." name=status value=in-progress
  4. obsidian task ref=<path:line> toggle to check off a DoD item
  5. obsidian property:set path="task/..." name=status value=done

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.

Check your setup, free