ai-workspace-lab/xworkspace-core-skills / skills/engineering-standards/agent-team-harness
agent-team-harness
VerifiedCoordinate multi-agent work through a tool-neutral contract for authoritative tasks, isolated execution, ownership, handoffs, evidence, acceptance, and archives. Use with harness-workflow for engineering delivery and load matching operations-management standards for operational effects.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 16.6 KB |
| references/field-flow.md | markdown | 2.7 KB |
| references/xworkspace-profile.example.json | config | 7.7 KB |
| references/xworkspace-profile.md | markdown | 6.2 KB |
| references/xworkspace-profile.schema.json | config | 11.3 KB |
references/field-flow.md
View on GitHub# Agent Team Harness field flow
This portable run card is subordinate to [Agent Team Harness](../SKILL.md), the selected TaskSource workflow, and linked engineering/operations standards.
## Intake
```text
[ ] Read repository instructions and confirm authorized scope.
[ ] Find the authoritative task and check for an existing duplicate.
[ ] Confirm goal, owner, business status, priority, and acceptance criteria in the TaskSource.
[ ] Record human-readable TASK_ID and immutable source ID (or explicit draft mapping).
[ ] Check active runs/ownership. If atomic claim/fencing is unavailable, serialize manually.
[ ] Bind run ID, executor/session, environment, isolated workspace/branch, and base revision.
[ ] Load harness-workflow and only the engineering/operations standards that apply.
```
## Source task comment
```text
Run: <run-id>
task_id: <stable-task-key>
Source Project node ID / item ID / issue node ID: <IDs-or-NOT_AVAILABLE>
Executor/environment: <non-sensitive labels>
Session binding: <approved reference; no token>
Workspace/branch/base revision: <workspace> / <branch> / <revision>
Evidence: <PR/artifact/check links and exact revision>
Next action or blocker: <bounded action or required decision>
```
## Handoff template
```markdown
# Handoff: <TASK_ID>
- Immutable source ID / relationships:
- Run ID / date:
- Prior run IDs (for recovery/takeover; preserve old records):
- Scope and explicit exclusions:
- Previous owner: stopped | active | unknown; evidence:
- Lease/fencing state and receiver confirmation:
- Executor/environment/session reference (non-sensitive):
- Workspace/branch/base revision/head revision/PR:
- Changed files and completed work:
- Checks: PASS | FAIL | BLOCKED | NOT_RUN (command, result, revision):
- Risks, unknown outcomes, and decisions needed:
- Next bounded action:
- Selected artifact IDs; archive: pending | file ID/URL/version/hash/idempotency key:
- Knowledge index: pending | task ID/source revision or archive ID/version/date/index reference:
```
## Closeout
```text
Engineering: PASS/FAIL/BLOCKED/NOT_RUN + exact revision/artifact
CI: PASS/FAIL/BLOCKED/NOT_RUN + run reference
Merge: PASS/FAIL/BLOCKED/NOT_RUN + PR/merge revision
Runtime: PASS/FAIL/BLOCKED/NOT_RUN + environment and boundary evidence
Business acceptance: authorized TaskSource owner decision
Archive: artifact ID + Drive/file ID + URL + version + integrity hash + source revision
Knowledge index: task ID + issue node ID + run ID + artifact ID + archive ref + QMD/index ref
Source task archive reference written: yes/no
Business completion authorized and applied: yes/no
```
Never mark missing evidence as passed. Resolve an uncertain write or dispatch by reading the target system before retrying.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |