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/context-locality.md
View on GitHub# Context Locality
Delegation exists to avoid forcing every phase, slice, review, command output, and compaction into one shrinking conversation.
## Principle
```text
Store broadly. Return compactly. Promote selectively. Load narrowly.
```
## Memory Tiers
- **Live context**: what the current agent sees now. Keep narrow.
- **Promoted context**: spec, plan, parent report, child result, handoff, decision note. Use for handoffs.
- **Recoverable context**: transcripts, events, raw tool output, routed output IDs. Store, but do not load by default.
- **Noise**: abandoned paths and verbose intermediates. Keep recoverable only when audit/debug value is worth it.
## Compression Points
- Researcher returns evidence summary, not transcript.
- Reviewer returns findings, not fixes.
- Worker returns changes/checks/uncertainty, not a session dump.
- Verifier returns command evidence and output IDs.
- Planning-parent returns `PLANNING_REPORT`.
- Execution-parent returns `EXECUTION_REPORT`.
## Artifact Role
Specs, plans, reviews, and handoffs are context compression artifacts. They are not ceremony. They preserve what future agents need without replaying the planning chat.
## Anti-Patterns
- Pasting child transcripts into parent context.
- Spawning agents because parallelism feels powerful.
- Giving every child the whole planning conversation.
- Treating recoverable history as always-loaded memory.
- Compacting one giant session repeatedly when phase/slice boundaries are available.
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 |