Skip to content

Gentleman-Programming/gentle-ai / internal/assets/skills/hermes-ephemeral-delegation

hermes-ephemeral-delegation

Verified

Trigger: a mapping need, parallel units, context backstop, high-risk verify, fresh review, or multi-step debug. Delegate via delegate_task for a named reason.

FileKindSize
SKILL.mdskill md4.5 KB
references/tuning-knobs.mdmarkdown1.3 KB

references/tuning-knobs.md

View on GitHub
# Hermes Delegation Tuning Knobs

Configure these in `~/.hermes/config.yaml` under the `delegation` key, or pass per `delegate_task` call.

| Parameter | Default | Effect |
|-----------|---------|--------|
| `max_spawn_depth` | 2 | Maximum recursive delegation depth. Set to 1 to prevent workers from spawning their own workers. |
| `max_concurrent_children` | 4 | Maximum number of parallel workers active at once. |
| `max_iterations` | agent default | Iteration budget for each worker before it is forced to return. |
| `child_timeout_seconds` | agent default | Hard wall-clock timeout per worker. |
| `inherit_mcp_toolsets` | false | When true, workers inherit the parent's MCP toolsets automatically. When false (default), pass toolsets explicitly in the mission. |
| `subagent_auto_approve` | false | When true, workers auto-approve all tool calls. When false (default), workers prompt for approval on dangerous calls. |

## Mission Checklist

When `inherit_mcp_toolsets` is false (the default), every mission passed to `delegate_task` MUST explicitly list:

- Which toolsets the worker is allowed to use (file read, shell, browser, etc.)
- Which MCP servers the worker can call (engram, context7, etc.)
- Which `SKILL.md` paths the worker must load before starting work

Without this, workers start with no tools beyond their built-in defaults.
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 322kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 322kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 322kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 198kUpdated Claude CodeCodexVerifiedInstall