Agent Work Receipts

Provides a safe, auditable AI coding workflow with repo briefs, logged commands, trustable receipts, and CI dry-run planning for transparent handoffs and rev...

Agent Work Receipts Skill

Use this meta-workflow when a coding task needs reliable orientation, command evidence, and human-readable proof.

Recommended flow

1. Preflight the repo

npx repo-agent-brief . > AGENT_BRIEF.md
sed -n '1,220p' AGENT_BRIEF.md

For branch handoffs:

npx repo-agent-brief . --diff origin/main > AGENT_HANDOFF.md

2. Run important commands with logs

npx agent-runlog -- npm test
npx agent-runlog -- npm run lint
npx agent-runlog -- npm run build

Use runlogs for noisy debugging and CI reproduction.

3. Produce final receipts

npx @builtbyecho/trustlog run -- npm test
npx @builtbyecho/trustlog verify .trustlog/latest.json

Use Trust Log as concise final evidence for the user, PR, issue, or next agent.

4. If CI is missing, preview before modifying

npx @builtbyecho/add-ci . --tier 2 --dry-run

Only rerun without --dry-run after the user/project owner agrees with the plan.

Decision guide

• Need orientation? repo-agent-brief.
• Need full command logs? agent-runlog.
• Need final proof/receipt? trustlog.
• Need CI scaffold planning? add-ci --dry-run.

Safety rules

• Check git identity before commits/pushes.
• Do not run destructive commands without explicit approval.
• Keep .agent-runs/, .trustlog/, AGENT_BRIEF.md, and AGENT_HANDOFF.md local unless the user wants them committed or shared.
• Inspect generated artifacts before public posting; redaction helps but is not a guarantee.