CtrlK
BlogDocsLog inGet started
Tessl Logo

jj-todo-workflow

Structured TODO commit workflow using JJ (Jujutsu). Use to plan tasks as empty commits with [task:*] flags, track progress through status transitions, manage parallel task DAGs with dependency checking. Enforces completion discipline. Enables to divide work between Planners and Workers. **Requires the working-with-jj skill**

88

1.50x
Quality

85%

Does it follow best practices?

Impact

96%

1.50x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

SKILL.md
Quality
Evals
Security

Evaluation results

98%

38%

Plan a Feature Rollout Using JJ

TODO DAG planning and description authoring

Criteria
Without context
With context

Uses --no-edit for creation

100%

100%

Uses jj-parallel-todos for branches

70%

100%

Correct todo flag on tasks

0%

100%

Title under 50 chars

62%

75%

Description has Context section

0%

100%

Description has Requirements section

0%

100%

Description has Acceptance criteria

100%

100%

No line-number references

100%

100%

Scripts invoked with path

0%

100%

Parallel merge uses multiple parents

100%

100%

draft vs todo distinction

100%

100%

plan-output.txt produced

100%

100%

100%

46%

Implement a Task in a JJ-Managed Repository

Task completion workflow and deviation documentation

Criteria
Without context
With context

Marks task wip on start

30%

100%

Uses jj-todo-next to review

50%

100%

Does NOT mark as done

100%

100%

Uses non-done status flag

0%

100%

Post-Implementation notes added

33%

100%

Deviation reason documented

80%

100%

Correct flag syntax used

0%

100%

jj-flag-update via script

50%

100%

workflow-output.txt produced

100%

100%

Implementation in src/validator.py

100%

100%

90%

12%

Coordinate Sub-Agent Work on a JJ Task Graph

Dependency checking and sequential agent coordination

Criteria
Without context
With context

Detects T2/T3/T4 as blocked

100%

100%

Uses jj log ancestors query

50%

50%

Sequential-by-default stated

40%

100%

Explains conflict reason

100%

100%

Workspace mechanism named

100%

100%

User approval required

100%

100%

Correct blocked flag usage

25%

25%

Recommended work sequence

100%

100%

Uses [task:*] flag format

62%

100%

check-deps-output.txt produced

100%

100%

Repository
YPares/agent-skills
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.