ledger-tasks-yylo
Skill511 starsChanged 4 months agoApache-2.0View on GitHub
Durable in-repo task state for coding agents — Kanban tasks with stable IDs, evidence-backed status marks, blocked-by dependencies, and next-ready selection. Use when starting, selecting, sequencing, or closing implementation tasks for a coding-agent project.
- Installs packages
2 other projects use a version of this.
Adopted by
Most recently taken up by
- cellrishi-code/claude-skills-atlastheir own version, first seen Oct 4, 2026
- Dianel555/DSkillstheir own version, first seen Sep 6, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- Dianel555/DSkills★ 65same name
- cellrishi-code/claude-skills-atlas★ 6same 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 repository32 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-browserSkill
- ai-elementsSkill
- autoresearchSkill
- axSkill
- better-iconsSkill
- bopomofo-rescueSkill
- browser-traceSkill
- codebase-designSkill
- deep-researchSkill
- diagnosing-bugsSkill
- domain-modelingSkill
- exa-searchSkill
- find-docsSkill
- find-skillsSkill
- get-api-docsSkill
- grillingSkill
- huashu-designSkill
- imagegen-frontend-webSkill
- improve-codebase-architectureSkill
- karpathy-guidelinesSkill
- kimi-webbridgeSkill
- migrate-to-shoehornSkill
- orca-replaySkill
- planning-with-filesSkill
- ponytailSkill
- product-decision-agentSkill
- prototypeSkill
- reverse-engineer-anythingSkill
- tddSkill
- teachSkill
- unslopSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 3 are distinctive, and 4 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.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use this skill when a coding agent needs durable, reviewable task state that lives in the repository instead of a chat transcript.
- Very specific: line 15
Start assigned work and record an evidence receipt (mark in_progress --response)
- Distinctive: line 26
If the task cannot proceed, keep the blocker explicit instead of closing it silently.
- Distinctive: line 51
Keep tasks small enough to finish in one iteration without filling the context window.
- Very specific: line 53
Always pass --response on mark.
- Very specific: line 54
Use ready before starting, and order --scores when planning parallel pipelines.
- Very specific: line 55
Read state with get before mutating.
≈ “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.
Files
YYLO Ledger Task Management
Use this skill when a coding agent needs durable, reviewable task state that lives in the repository instead of a chat transcript. YYLO Ledger is the source of truth for task state; this skill teaches the working loop.
Adapted from the canonical ledger-tasks-yylo skill in yylo-dev/yylo-skills; see the upstream SKILL.md for the full CLI reference (archives, multi-directory merges, controller routing).
When to Use
- Decide which task is safe to start next (
ready= all blockers done) - Start assigned work and record an evidence receipt (
mark in_progress --response) - Close a task with what changed and how it was tested (
mark done --response --commit) - Declare dependencies between tasks before creating them
- Sequence parallel work topologically (
order --scores)
Workflow
- Pick work:
yy ledger ready(oryy ledger list --status todo,in_progress). - Inspect full state before any mutation:
yy ledger get TASK_ID(includes resolvedblocked_byand related tasks). - Start:
yy ledger mark in_progress --id TASK_ID --response "Starting: <plan>". - Finish:
yy ledger mark done --id TASK_ID --response "Done: <what> — tested <how>" --commit <sha>. - If the task cannot proceed, keep the blocker explicit instead of closing it silently.
Core Commands
yy ledger create "Short, one-iteration task" --status backlog --tags feature,backend
yy ledger list --status todo,in_progress --limit 10
yy ledger search --tag backend --open
yy ledger get TASK_ID
yy ledger update TASK_ID --tags backend,urgent
yy ledger deps add --id TASK_ID --blocked-by BLOCKER1 BLOCKER2
yy ledger ready --tag backend
yy ledger order --scores
yy ledger archive TASK_ID # soft delete, preserves dataInline body markup is parsed on create/update, so dependencies can be declared in prose:
[blocked_by]TASK_ID[/blocked_by] — this task waits on TASK_ID
[task_id]RELATED_ID[/task_id] — cross-reference a related taskBest Practices
- Keep tasks small enough to finish in one iteration without filling the context window.
- Status flow:
backlog -> todo -> in_progress -> done(archivefor abandoned tasks). - Always pass
--responseonmark; attach--commitwhen marking done so task history links to the diff. - Use
readybefore starting, andorder --scoreswhen planning parallel pipelines. - Read state with
getbefore mutating; never bypass the CLI with direct file edits.
Prerequisites
- YYLO CLI:
npm install --global '@yylo/cli@latest'(Node.js 20.10+, npm, Git).yy ledgerdelegates to the standalone YYLO Ledger Git-native task store (MIT). - Task state lives under the repository's
.juno_task/directory and is committed with the project, so any agent or reviewer can reconstruct decisions from history.
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
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.

