hassan-mohiddin/freeflow / .deprecated/delegation-harness/skills/delegation-harness
delegation-harness
VerifiedUse when coordinating Freeflow Pi/cmux pane delegation, orchestrator/planning-parent/execution-parent workflows, task packets, child results, context locality, capability reroutes, multi-agent execution, work packages, worktrees, routing fresh-reviewer/subagent requests, or when the user asks to spawn/manage visible pane agents.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 12.0 KB |
| references/context-locality.md | markdown | 1.5 KB |
| references/execution-and-integration.md | markdown | 3.9 KB |
| references/reviewer-profiles.md | markdown | 3.8 KB |
| references/roles-and-contracts.md | markdown | 4.1 KB |
| references/task-packets-and-results.md | markdown | 5.1 KB |
| references/tool-policy.md | markdown | 1.7 KB |
references/reviewer-profiles.md
View on GitHub# Reviewer Profiles
Use profiles to select a review perspective, not to grant authority or prescribe a model. The harness chooses the available agent, fresh context, transport, and tools.
Reviewer profiles receive the applicable Freeflow reviewer contract, source truth, work product, direct evidence, pass history, and residual risk. The independent verifier receives the verifier contract and finalized evidence package instead; it is not a reviewer profile. Domain prestige does not replace evidence.
## Requirements Reviewer
Focus:
- accepted outcome, scope, and non-goals;
- missing, extra, or misunderstood behavior;
- public and failure contracts;
- source-truth conflicts and hidden owner decisions.
Use for specs, plans, task completion, and integration boundaries.
## Code-Quality Reviewer
Focus:
- correctness and regression risk;
- readability and unnecessary complexity;
- module depth, locality, and contract surface;
- test quality and maintainability;
- performance/security concerns only where the diff creates a concrete path.
Do not use “would a staff engineer approve” as evidence. Cite the requirement and consequence.
## Independent Verifier
The verifier is a distinct fresh context from implementer and reviewer. It runs the finalized allowed checks once against the unchanged final state and reports:
- exact state and environment identity;
- commands, exit status, and evidence pointers;
- whether checks exercised the claimed seam;
- proved, unproved, reduced-fidelity, failed, or unavailable claims;
- unexpected mutations or side effects.
It does not inspect broad design quality, decide product behavior, edit, or fix. It may run in parallel with a distinct reviewer against the same frozen state without sharing outputs. If another independent run is needed, the parent must obtain user authorization.
A test-quality reviewer is different: it may judge whether tests are representative or over-mocked, but its judgment does not count as independent verification.
## Security Reviewer
Use when trust boundaries or sensitive behavior change. Load `../../review-work/references/security-risk-lens.md` and applicable repo/domain policy.
Focus on concrete assets, actors, paths, controls, failure behavior, and evidence. Escalate specialized cryptography, identity, sandbox, or infrastructure questions rather than guessing.
## Performance Reviewer
Use when requirements or evidence include latency, throughput, memory, capacity, client responsiveness, or regression risk.
Focus on representative baseline, measurement method, profiler/query-plan/trace evidence, variance, correctness tradeoffs, and whether the claimed improvement survives the real path. Avoid universal budgets.
## Accessibility Reviewer
Use for affected user-interface paths when accessibility requirements apply.
Use the repo/design-system authority and available domain tooling. Focus on keyboard/focus, semantics, names/roles/states, assistive-technology behavior, visual constraints, and loading/error/degraded states relevant to the change.
## Release / Operations Reviewer
Use for releases, migrations, production launch, rollback, or operational readiness.
Focus on artifact/config identity, compatibility, data safety, observability, advance/abort criteria, rollback or forward recovery, temporary machinery, and consumer/operator evidence.
## Composition
Use one reviewer with the fewest material lenses that cover the boundary. Multiple reviewers for one boundary are additional dispatches and require scoped user authorization; do not parallelize reviewers merely to increase confidence or finding count.
Do not count votes. The parent adjudicates findings against source truth. One narrow confirmation requires scoped authorization and accepted blockers; a third pass is exceptional, owner-selected, and terminal.
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: 323k | 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: 323k | 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: 323k | Updated | Claude CodeCodex | Verified | Install |