Human Checkpoint Agent Skill

Human Checkpoint

A checkpoint is consent, not a formatting step. It binds a named human

Decision to the exact Artifacts and constraints reviewed before execution may

continue.

Inputs

• DECISIONS.md.
• UNITS.csv and STATUS.md for the active checkpoint.
• Artifacts declared by the locked Pipeline for that checkpoint.

Outputs

• Updated DECISIONS.md.
• A checkpoint Decision in the Run ledger.

Steps

1. Identify the pending checkpoint

Inspect STATUS.md, UNITS.csv, and the active runner message. Select the

first blocked HUMAN Unit and its C* identifier. If multiple checkpoints

appear active or the checklist is missing, stop and repair the projection

before approving anything.

Completion criterion: exactly one pending checkpoint and its owning HUMAN Unit

are identified.

2. Review the declared Artifacts

Read the locked Pipeline's checkpoint contract and inspect every named Artifact.

Record requested constraints or scope changes in the checkpoint block before

approval; do not silently modify reader-facing content as part of sign-off.

Completion criterion: the reviewer can name the Artifacts inspected and any

constraints attached to the Decision.

3. Record approval through the adapter

Use the Pipeline adapter so the Markdown checkbox and machine Decision ledger

remain synchronized:

uv run python scripts/pipeline.py approve \
  --workspace workspaces/<name> \
  --checkpoint <C*>

Do not infer approval from chat silence, a completed Artifact, or an existing

but unchecked checklist item.

Completion criterion: DECISIONS.md contains [x] Approve C* and the Run

ledger records checkpoint.approved for the same checkpoint.

4. Hand execution back to the Runner

Resume through the Pipeline adapter. The Harness may complete the HUMAN Unit

and expose the next eligible Unit; this Skill does not execute downstream

semantic work itself.

Completion criterion: the checkpoint is no longer the active blocker, or one

new specific blocker is visible in durable Workspace state.

Context Pointers

• The locked pipelines/*.pipeline.md owns checkpoint purpose and required

review Artifacts.

• DECISIONS.md is the human-readable Decision surface.
• .harness/decisions.jsonl is the machine-readable history; update it

through the adapter rather than by hand.

• Use checkpoint-brief to recreate a post-route checkpoint review block.
• scripts/run.py is a runner-compatibility helper that only toggles the

Markdown checkbox. Prefer scripts/pipeline.py approve, which also records

the machine Decision.

Troubleshooting

• If the approvals checklist is missing after C0, materialize the review block

with checkpoint-brief before approval. Use pipeline-router only for the

initial C0 route.

• If reviewed upstream Artifacts later change, expect the Harness to revoke the

stale approval and request a new Decision.