Activation Contract
Load this skill when you are acting as the parent orchestrator and the work ahead falls into any of these categories:
- Broad exploration (4+ files to understand, codebase mapping, approach comparison)
- Multi-file implementation (touching 2+ non-trivial files)
- Test or build execution
- Fresh adversarial review (diffs, PR readiness, incident audit)
- Multi-step debugging that would flood the parent context
Do NOT load this skill if you are already inside a delegated child task — you are the executor, not the orchestrator.
Hard Rules
- Use
delegate_task for all complex work listed above. Do NOT execute it inline.
- Workers are EPHEMERAL: each
delegate_task call creates a fresh context. Do NOT request persistent agent files or profiles.
- Pass a self-contained mission. Workers have no memory of the parent conversation.
- Treat worker output as self-report: verify file writes, test pass/fail, URLs, and IDs before reporting success to the user.
- Batch parallel calls only for INDEPENDENT workstreams. Sequential dependencies must run sequentially.
Decision Gates
| Situation |
Action |
| Need to read 4+ files to understand |
Delegate a narrow exploration worker |
| Need to write 2+ non-trivial files |
Delegate a single writer with the full mission |
| Need to run tests or builds |
Delegate an executor; do not run inline |
| Need an adversarial review of a diff |
Delegate a fresh-context reviewer |
| Multi-step debug that grows the context |
Delegate a debug worker; feed results back inline |
| Simple 1-file edit you already understand |
Do it inline; no delegation needed |
| Quick git/state check |
Do it inline; no delegation needed |
Execution Steps
- Identify which gate applies. If none applies, skip delegation.
- Draft a self-contained mission for the worker — include:
- Exact goal (one sentence)
- File paths or targets to act on
- Relevant prior context the worker needs (decisions, conventions, prior findings)
- Constraints (style, test runner, budget)
- Expected evidence to return (e.g., file written, test output, URL found)
- Allowed toolsets/MCP/skills the worker should use
- Any
SKILL.md paths to load before work
- Call
delegate_task with that mission.
- Wait for the worker summary.
- Verify the claimed output (check file existence, test result, side effect).
- Synthesize the verified result into your orchestrator reply.
Output Contract
After synthesizing worker results, return:
- What was delegated and to how many workers
- What each worker returned (verified, not just claimed)
- Any discrepancy between worker self-report and verified evidence
- Final answer or next step for the user
References
- references/tuning-knobs.md — Full table of
delegate_task configuration parameters and the explicit toolset/MCP/skill checklist for worker missions.
- ../../hermes/sdd-orchestrator.md — SDD orchestrator protocol that uses this delegation standard for SDD phase work.
Source: Gentleman-Programming/gentle-ai — distributed by TomeVault.
1---2name: gentleman-programming-gentle-ai-gentle-ai3description: Activation Contract4---56## Activation Contract78Load this skill when you are acting as the parent orchestrator and the work ahead falls into any of these categories:910- Broad exploration (4+ files to understand, codebase mapping, approach comparison)11- Multi-file implementation (touching 2+ non-trivial files)12- Test or build execution13- Fresh adversarial review (diffs, PR readiness, incident audit)14- Multi-step debugging that would flood the parent context1516Do NOT load this skill if you are already inside a delegated child task — you are the executor, not the orchestrator.1718## Hard Rules1920- Use `delegate_task` for all complex work listed above. Do NOT execute it inline.21- Workers are EPHEMERAL: each `delegate_task` call creates a fresh context. Do NOT request persistent agent files or profiles.22- Pass a self-contained mission. Workers have no memory of the parent conversation.23- Treat worker output as self-report: verify file writes, test pass/fail, URLs, and IDs before reporting success to the user.24- Batch parallel calls only for INDEPENDENT workstreams. Sequential dependencies must run sequentially.2526## Decision Gates2728| Situation | Action |29|-----------|--------|30| Need to read 4+ files to understand | Delegate a narrow exploration worker |31| Need to write 2+ non-trivial files | Delegate a single writer with the full mission |32| Need to run tests or builds | Delegate an executor; do not run inline |33| Need an adversarial review of a diff | Delegate a fresh-context reviewer |34| Multi-step debug that grows the context | Delegate a debug worker; feed results back inline |35| Simple 1-file edit you already understand | Do it inline; no delegation needed |36| Quick git/state check | Do it inline; no delegation needed |3738## Execution Steps39401. Identify which gate applies. If none applies, skip delegation.412. Draft a self-contained mission for the worker — include:42 - Exact goal (one sentence)43 - File paths or targets to act on44 - Relevant prior context the worker needs (decisions, conventions, prior findings)45 - Constraints (style, test runner, budget)46 - Expected evidence to return (e.g., file written, test output, URL found)47 - Allowed toolsets/MCP/skills the worker should use48 - Any `SKILL.md` paths to load before work493. Call `delegate_task` with that mission.504. Wait for the worker summary.515. Verify the claimed output (check file existence, test result, side effect).526. Synthesize the verified result into your orchestrator reply.5354## Output Contract5556After synthesizing worker results, return:5758- What was delegated and to how many workers59- What each worker returned (verified, not just claimed)60- Any discrepancy between worker self-report and verified evidence61- Final answer or next step for the user6263## References6465- [references/tuning-knobs.md](references/tuning-knobs.md) — Full table of `delegate_task` configuration parameters and the explicit toolset/MCP/skill checklist for worker missions.66- [../../hermes/sdd-orchestrator.md](../../hermes/sdd-orchestrator.md) — SDD orchestrator protocol that uses this delegation standard for SDD phase work.6768---69> Source: [Gentleman-Programming/gentle-ai](https://github.com/Gentleman-Programming/gentle-ai) — distributed by [TomeVault](https://tomevault.io).70<!-- tomevault:4.0:skill_md:2026-06-27 -->
tomevault-io