Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Approval-Checkpoint Long Task · mkurman | DecimalAI

agent toolingby mkurman

Approval-Checkpoint Long Task

mkurman/approval-checkpoint-long-task

Source mkurman/zorai ↗License MIT

Canonical long-task pack for daemon-managed work with deliberate approval checkpoints, status summaries, rollback notes, and mobile-safe governance-aware updates.

SkillScore87Maintained

Automatic

Project Management pack

Last verified 8/21/2026

What it changes

Passed 20 of 23 test cases with the skill, and 10 of 23 without it.

Without skill43%

With skill87%

Checks it passed that the model alone missed

Includes task_kind specifying task or goal.
Reports rollback notes before attempting any step retry.
Configures the task_kind input field to select between task and goal.

and 12 more in the full report

10 fixed10 passed either way3 still failing0 broke

Pass-rate lift: +43 pp
Tokens: -20%
Agent turns: 0%
Response time: -47%

Model gemini-3.6-flashMeasured 8/21/2026v120 of 23 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/mkurman-approval-checkpoint-long-task/SKILL.md https://app.decimal.ai/s/mkurman-approval-checkpoint-long-task/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add mkurman-approval-checkpoint-long-task, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/mkurman-approval-checkpoint-long-task/ and pins it in .decimal/skills.lock

decimalai skills add mkurman-approval-checkpoint-long-task

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/mkurman-approval-checkpoint-long-task/SKILL.md https://app.decimal.ai/s/mkurman-approval-checkpoint-long-task/SKILL.md

Raw file: /s/mkurman-approval-checkpoint-long-task/SKILL.md ↗ · machine-readable summary at /s/mkurman-approval-checkpoint-long-task.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 10

Prompt

We are configuring a multi-step workflow. One scenario has a fixed set of execution steps, while another scenario seeks to achieve an objective with fluid sub-steps. What input field controls this primitive distinction?

What was checked, without and with the skill
Checked	Without skill	With skill
Configures the task_kind input field to select between task and goal.	Missed	Passed

case-11 · gemini-3.6-flash · 8/21/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md57 linessingle file

Approval-Checkpoint Long Task

User story

I want long-running work to pause at explicit checkpoints with clear status, next step, and rollback notes, so I can approve or deny risky transitions from in-app or mobile chat without losing context.

Pack contract

Prerequisites and readiness

Reuses daemon task or goal primitives only
Chat/mobile approval surfaces are optional but preferred for remote control
If remote approval channels are unavailable, checkpoints must still work in-app

Inputs and configuration fields

task_kind: task or goal
checkpoint_titles: ordered checkpoint labels
rollback_notes: per-checkpoint rollback instructions
summary_cadence: when to emit status summaries

Outputs and delivery targets

current checkpoint
completed checkpoints
next action
rollback state / notes
approval state and recovery guidance

Manual run recipe

Start daemon task/goal with explicit checkpoint plan in the description.
Before each risky transition, summarize state and rollback notes.
Pause for approval.
Resume only after valid approval resolution.

Example routine wiring

This pack is usually manual or task/goal-backed rather than purely cron-driven; if scheduled, it should only materialize the governed task template, not auto-bypass approvals.

Track record

Benchmark

20 / 23

test cases passed, vs 10 / 23 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

mkurman / zorai

skills/workflow-packs/approval-checkpoint-long-task/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Approval-Checkpoint Long Task

User story

Pack contract

Prerequisites and readiness

Inputs and configuration fields

Outputs and delivery targets

Manual run recipe

Example routine wiring

Track record

Benchmark

Adoption

SkillSafety

Source

Related skills