vitya
38ac9ef5d1
Un-reds the hermes build (build fails on unmapped skills in skills/; this skill landed in sources 2026-06-16 unmapped). Entry is mode:pending, intended auto/meta — a purely behavioral governance skill (no tool-side effects) that should pass a non-implementer test-trigger + review before auto-firing. Filed ⚪ inter-session-peer-discipline-test-trigger + 🔵 -review. Zone confirmed claude-skills (skill lives in this repo). Human-ratified; workshop coordinated as a peer proposal, not authority (per the skill itself). YAML validated (33 skills, 9 pending). Schema version untouched. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
205 lines
8.1 KiB
YAML
205 lines
8.1 KiB
YAML
# Hermes Skills Rollout — mapping
|
|
#
|
|
# Source of truth for the converter (scripts/build-hermes.py).
|
|
# EVERY skill in skills/ MUST appear in `skills:` exactly once — the build
|
|
# fails on unmapped skills (forces an explicit decision per skill).
|
|
#
|
|
# Modes:
|
|
# auto — convert from skills/<name>/, applying replace-rules. Lands in
|
|
# dist-hermes/<category>/<name>/.
|
|
# manual — copy from `source:` directory verbatim (no replace-rules).
|
|
# Used for Hermes-flavour rewrites (e.g. yaml-edit MCP setups).
|
|
# skip — audited as not-applicable; recorded in dist-hermes/SKIPPED.md.
|
|
# pending — present in mapping but not yet built; recorded in SKIPPED.md.
|
|
# `intended:` block declares the future mode/category, so the
|
|
# audit table is preserved across MVP iterations.
|
|
#
|
|
# Categories: software-development | productivity | mcp | research | meta
|
|
#
|
|
# See .wiki/concepts/hermes-skills-rollout-design.md for the full audit
|
|
# (Q1-Q8 + the 21-row decision table).
|
|
|
|
version: 1
|
|
|
|
hermes:
|
|
installer_path: meta/claude-skills-installer
|
|
default_category: software-development
|
|
|
|
categories:
|
|
- software-development
|
|
- productivity
|
|
- mcp
|
|
- research
|
|
- meta
|
|
|
|
skills:
|
|
|
|
# ─── auto (4 universals — MVP target) ────────────────────────────────
|
|
|
|
pulling-before-work:
|
|
mode: auto
|
|
category: software-development
|
|
|
|
active-platform:
|
|
mode: auto
|
|
category: software-development
|
|
replace-rules:
|
|
- find: "**Windows + PowerShell.**"
|
|
replace: "**Linux + bash.**"
|
|
reason: "Hermes runs on Linux factory machines (/opt/data/projects). Default platform shifts."
|
|
- find: "Use this until something in the session tells you to switch. Reason: the user's primary workstation is Windows."
|
|
replace: "Use this until something in the session tells you to switch. Reason: Hermes runs on Linux factory machines."
|
|
reason: "Mirror reasoning sentence to match the new default."
|
|
|
|
project-discipline:
|
|
mode: auto
|
|
category: software-development
|
|
|
|
tdd-criteria:
|
|
mode: auto
|
|
category: software-development
|
|
|
|
using-markitdown:
|
|
mode: auto
|
|
category: productivity
|
|
|
|
# ─── skip (8 — audited as not-applicable) ────────────────────────────
|
|
|
|
caveman:
|
|
mode: skip
|
|
reason: "Hermes runs on glm-5.1 (cheap local model); token-compression motive disappears."
|
|
caveman-commit:
|
|
mode: skip
|
|
reason: "Caveman family — cheap-model context, no compression motive."
|
|
caveman-compress:
|
|
mode: skip
|
|
reason: "Caveman family — cheap-model context, no compression motive."
|
|
caveman-help:
|
|
mode: skip
|
|
reason: "Help-card for the caveman family which itself is skipped."
|
|
caveman-review:
|
|
mode: skip
|
|
reason: "Caveman family — cheap-model context, no compression motive."
|
|
setup-interns:
|
|
mode: skip
|
|
reason: "Hermes itself is a cheap-intern model — the delegation tier collapses."
|
|
using-interns:
|
|
mode: skip
|
|
reason: "Hermes itself is a cheap-intern model — the delegation tier collapses."
|
|
find-skills:
|
|
mode: skip
|
|
reason: "Hermes has built-in skills_list() / progressive disclosure."
|
|
|
|
# ─── pending (10 — deferred to follow-up tasks) ──────────────────────
|
|
|
|
setup-tasks:
|
|
mode: auto
|
|
category: productivity
|
|
|
|
using-tasks:
|
|
mode: auto
|
|
category: productivity
|
|
|
|
setup-wiki:
|
|
mode: auto
|
|
category: research
|
|
|
|
using-wiki:
|
|
mode: auto
|
|
category: research
|
|
|
|
setup-projects-meta:
|
|
mode: manual
|
|
source: hermes/skills/setup-projects-meta
|
|
category: mcp
|
|
|
|
using-projects-meta:
|
|
mode: auto
|
|
category: mcp
|
|
|
|
setup-context7:
|
|
mode: manual
|
|
source: hermes/skills/setup-context7
|
|
category: mcp
|
|
|
|
using-context7:
|
|
mode: auto
|
|
category: mcp
|
|
|
|
project-bootstrap:
|
|
mode: auto
|
|
category: software-development
|
|
|
|
recommend-dont-menu:
|
|
mode: auto
|
|
category: productivity
|
|
|
|
update-claude-skills:
|
|
mode: skip
|
|
reason: "Claude-Code-only orchestrator — Hermes uses hermes-installer-skill instead."
|
|
|
|
# ─── pending (9 — behavioral audit required) ─────────────────────────
|
|
|
|
delegate-task:
|
|
mode: pending
|
|
intended:
|
|
mode: auto
|
|
category: mcp
|
|
reason: "Calls mcp__projects-meta__tasks_create to create tasks in other projects/agents (Gitea commit, cross-project side-effect). Behavioral audit via delegate-task-test-trigger required before promotion to auto."
|
|
|
|
using-yt-tools:
|
|
mode: pending
|
|
intended:
|
|
mode: auto
|
|
category: research
|
|
reason: "Shells out to yt-dlp + ffmpeg and writes ./yt-cache/ in cwd. Behavioral audit via using-yt-tools-test-trigger required before promotion to auto."
|
|
|
|
using-vds-ops:
|
|
mode: pending
|
|
intended:
|
|
mode: auto
|
|
category: mcp
|
|
reason: "Calls mcp__vds-ops__* tools (read-only, but touches infrastructure). Behavioral audit via using-vds-ops-test-trigger required before promotion to auto."
|
|
|
|
using-wiki-graph:
|
|
mode: pending
|
|
intended:
|
|
mode: auto
|
|
category: mcp
|
|
reason: "Calls mcp__wiki-graph__* tools (read-only, parses a .wiki/ corpus server-side). Behavioral audit via using-wiki-graph-test-trigger required before promotion to auto."
|
|
|
|
session-handoff:
|
|
mode: pending
|
|
intended:
|
|
mode: auto
|
|
category: productivity
|
|
reason: "Writes .tasks/NEXT_SESSION.md (project-scope, sliding overwrite) and reads it on session start. Bidirectional file-system side-effect, opt-in via CLAUDE.md trigger-line. Behavioral audit via session-handoff-test-trigger required before promotion to auto."
|
|
|
|
private-dev-public-publish:
|
|
mode: pending
|
|
intended:
|
|
mode: auto
|
|
category: software-development
|
|
reason: "Steps shell out to git / gh / Gitea-API, handle tokens, force-push, and repo deletion/privacy toggles — not a purely stylistic skill. Behavioral audit via private-dev-public-publish-test-trigger required before promotion to auto."
|
|
|
|
task-loop:
|
|
mode: pending
|
|
intended:
|
|
mode: auto
|
|
category: mcp
|
|
reason: "Orchestrates the board claim/close/update/heartbeat cycle via mcp__projects-meta__tasks_claim_next / tasks_close / tasks_update / tasks_heartbeat (cross-session claim ownership, irreversible close, Gitea side-effects) and may arm a single long ScheduleWakeup for the explicit long-watch opt-in. Critical-infra-adjacent — touches the same claim/close machinery the unattended poller relies on. Behavioral audit via task-loop-test-trigger required before promotion to auto."
|
|
|
|
session-inbox-monitor:
|
|
mode: pending
|
|
intended:
|
|
mode: auto
|
|
category: productivity
|
|
reason: "Paired SessionStart hook registers itself in ~/.claude/settings.json and sweeps orphaned monitor OS processes (Get-CimInstance | Stop-Process by sentinel+inbox-path); the skill then raises an in-session Monitor on .claude-inbox/. Machine-level side-effects (user-config mutation + process kills) and a Windows-PowerShell hook that needs a Linux port for Hermes factory machines. Primary activation is the CLAUDE.md trigger-line `inbox monitor: raise on start` + the injector, not a hermes-trigger. Behavioral audit via session-inbox-monitor-test-trigger required before promotion to auto."
|
|
|
|
inter-session-peer-discipline:
|
|
mode: pending
|
|
intended:
|
|
mode: auto
|
|
category: meta
|
|
reason: "Purely behavioral governance skill — NO tool-side effects (no settings.json write, no process kill, no Monitor raise). Governs how an agent treats inter-session/peer messages: peer messages (and the agent's own replies) are proposals/analysis, NOT authority; the human is the only source of scope/direction; a peer-driven design escalation must not be reported as settled without explicit human ratification. Pending (not auto) by the same caution applied to other unverified behavioral skills: a behavioral-authority skill should not auto-fire until it passes a non-implementer test-trigger + review. Behavioral audit via inter-session-peer-discipline-test-trigger required before promotion to auto. (Ledger filed 2026-06-17 per workshop coordination; human-ratified.)"
|