Skip to main content
The build command assigns one builder subagent to implement the task against the frozen spec.md. After implementation, the same subagent session is the preferred place to pack evidence — this is the same-session evidence packing pattern. Only run a separate evidence-only session if the original builder session is unavailable.
The builder must not write verdict.json or problems.md at any point, and must not claim final task completion.

​
BUILD prompt

Copy this prompt verbatim when delegating to the builder subagent. Replace <TASK_ID> with the actual task identifier. 
You are in BUILD mode for TASK_ID <TASK_ID>.

Read:
- .agent/tasks/<TASK_ID>/spec.md
- AGENTS.md if present
- CLAUDE.md if present
- .claude/CLAUDE.md if present
- relevant .claude/rules/**/*.md files if present

Your job:
- Implement the task against the frozen spec
- Make the smallest safe change set that satisfies the acceptance criteria
- Run focused checks as needed
- Keep unrelated files untouched
- Do not write verdict.json or problems.md
- Do not claim final completion yet

Return to the parent with:
- files changed
- checks run
- open risks

​
What the builder reads and implements

  • .agent/tasks/<TASK_ID>/spec.md — the frozen acceptance criteria and verification plan
  • AGENTS.md if present
  • CLAUDE.md if present
  • .claude/CLAUDE.md if present
  • Relevant .claude/rules/**/*.md files if present
The smallest safe change set in production code that satisfies every acceptance criterion in spec.md. Unrelated files must remain untouched.
  • List of files changed
  • Checks run during implementation
  • Open risks or remaining uncertainties

​
Same-session evidence packing

After the builder finishes implementation, continue in the same subagent session with the EVIDENCE follow-up prompt below. This is the default path in Claude Code.
Only use the EVIDENCE-ONLY fallback when the original builder session is unavailable or you intentionally want a fresh evidence-only run.

​
EVIDENCE follow-up prompt

Send this as a follow-up message to the same builder session immediately after BUILD completes. 
PACK EVIDENCE for TASK_ID <TASK_ID>.

Do not change production code.

Read:
- .agent/tasks/<TASK_ID>/spec.md
- the current repository state
- any prior command results from this builder session

Write or update:
- .agent/tasks/<TASK_ID>/evidence.md
- .agent/tasks/<TASK_ID>/evidence.json
- .agent/tasks/<TASK_ID>/raw/build.txt
- .agent/tasks/<TASK_ID>/raw/test-unit.txt
- .agent/tasks/<TASK_ID>/raw/test-integration.txt
- .agent/tasks/<TASK_ID>/raw/lint.txt
- .agent/tasks/<TASK_ID>/raw/screenshot-1.png when a screenshot is useful

Rules:
- For each AC, assign PASS, FAIL, or UNKNOWN
- Every PASS must cite concrete proof
- FAIL and UNKNOWN must explain the gap
- Overall PASS only if every AC is PASS
- Prefer raw artifacts over narrative prose

Return only:
- overall_status
- created or updated files
- commands a fresh verifier should rerun

​
EVIDENCE-ONLY fallback prompt

Use this prompt only when the original builder session is unavailable. 
You are in EVIDENCE-ONLY mode for TASK_ID <TASK_ID>.

Read:
- .agent/tasks/<TASK_ID>/spec.md
- the current repository state

Write the same evidence bundle as above.

Do not change production code.

​
Hard boundaries

BUILD mode

Implement against the spec. Must not write verdict.json or problems.md.

EVIDENCE mode

Do not change production code. Pack artifacts only.

Build docs developers (and LLMs) love

Get started for freeTalk to us