Skip to content

feat(agents): add CMTrace Open specialist agent skill + Clairvoyance staff org - #516

Merged
adamgell merged 3 commits into
mainfrom
feat/cmtraceopen-agent-skill
Aug 7, 2026
Merged

feat(agents): add CMTrace Open specialist agent skill + Clairvoyance staff org#516
adamgell merged 3 commits into
mainfrom
feat/cmtraceopen-agent-skill

Conversation

@adamgell

@adamgell adamgell commented Aug 7, 2026

Copy link
Copy Markdown
Owner

What

Adds a complete agent skill system for CMTrace Open:

Agent Identity Files

  • soul.md — Agent identity, 8 operating rules, model tiering, decision framework, verified checkpoints
  • memory.md — Durable facts: architecture, checkpoints with SHAs, recovery branches, execution order, 450-worktree ecosystem

Hermes Skill

  • .agents/skills/cmtraceopen/SKILL.md — Thin wrapper skill that points to repo files as canonical source

Clairvoyance Staff Org (Rebuilt)

Model Tiering

  • Scaffold: kimi-k2.7-code, deepseek-v4-flash, qwen-flash, gpt-5-luna
  • Mid: kimi-k3, grok-4-20-reasoning
  • Reasoning: gpt-5.6-sol, claude-opus-4-8

Sources

All content grounded in:

  • ~/.hermes/cmtrace-pm-charter.md (execution manager contract)
  • AGENTS.md (simplicity rules)
  • CLAUDE.md (build commands, architecture)
  • 450-worktree ecosystem state
  • Session history with Codex/Claude agent work

No fabrication — every claim traces to a real source.

Summary by CodeRabbit

  • Documentation
    • Added comprehensive project guidance covering architecture, workflows, quality standards, parser boundaries, evidence requirements, and operating rules.
    • Added role charters for leadership, engineering, technical writing, and UI/design responsibilities.
    • Added navigation indexes for repository references, shared memory, staff notes, and documentation review tasks.
    • Added onboarding guidance for the project-specialist agent, including startup instructions and key resources.
    • Added an assessment kickoff prompt defining review activities, reporting expectations, and implementation restrictions.

- Add soul.md: agent identity, operating rules, model tiering, decision framework
- Add memory.md: durable facts, verified checkpoints, recovery branches, execution order
- Add .agents/skills/cmtraceopen/SKILL.md: Hermes skill entry point referencing repo files

Built from PM charter, Clairvoyance staff org, AGENTS.md, CLAUDE.md, and
450-worktree ecosystem state. No fabrication — all content grounded in
real repo sources and Adam's documented workflow.
- Route to soul.md and memory.md for CMTrace Open specialist agent loading
…fold tier

- Rebuild .Clairvoyance/staff/ with ceo, coder, ui-design, tech-writer charters
- Add Roger and Theo staff member directories with notes
- Add .Clairvoyance/memory/index.md for shared workspace memory
- Add .Clairvoyance/library.md routing index with agent skill links
- Add .Clairvoyance/kickoff-prompt.md for CEO first message
- Update all model tiering to include gpt-5-luna in scaffold rotation
- Update soul.md, memory.md, and SKILL.md with corrected tiers
Copilot AI lite review requested due to automatic review settings August 7, 2026 00:54
@github-actions github-actions Bot added enhancement New feature or request feature New feature labels Aug 7, 2026
@coderabbitai

coderabbitai Bot commented Aug 7, 2026

Copy link
Copy Markdown

Review Change Stack

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info
⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 16b5d242-a4d6-4692-afe4-bb1223bd95fb

📥 Commits

Reviewing files that changed from the base of the PR and between 4eceb5d and 9d62d12.

📒 Files selected for processing (13)
  • .Clairvoyance/kickoff-prompt.md
  • .Clairvoyance/library.md
  • .Clairvoyance/memory/index.md
  • .Clairvoyance/staff/ceo-charter.md
  • .Clairvoyance/staff/coder-charter.md
  • .Clairvoyance/staff/roger/index.md
  • .Clairvoyance/staff/tech-writer-charter.md
  • .Clairvoyance/staff/theo/index.md
  • .Clairvoyance/staff/ui-design-charter.md
  • .agents/skills/cmtraceopen/SKILL.md
  • library.md
  • memory.md
  • soul.md

📝 Walkthrough

Walkthrough

Added the CMTrace Open agent framework. It defines project context, operating rules, specialist loading, staff charters, routing indexes, durable memory, and a CEO assessment kickoff process.

Changes

CMTrace Open agent framework

Layer / File(s) Summary
Project context and operating rules
soul.md, memory.md
Added project identity, architecture facts, evidence rules, workflow procedures, validation commands, checkpoints, staff roles, model tiers, and governing rules.
Specialist skill and workspace loading
.agents/skills/cmtraceopen/SKILL.md, library.md
Added the cmtraceopen specialist skill and instructions to load soul.md and memory.md.
Staff charters and note routing
.Clairvoyance/staff/*
Added CEO, Coder, Technical Writer, and UI/design charters. Added Roger and Theo note-routing indexes.
Workspace and memory navigation
.Clairvoyance/library.md, .Clairvoyance/memory/index.md
Added topic routing, WikiLink rules, repository path references, and memory-note routing instructions.
CEO assessment kickoff
.Clairvoyance/kickoff-prompt.md
Added assessment instructions for staffing, governing documents, checkpoints, blockers, reporting, and restricted repository activity.

Estimated code review effort: 2 (Simple) | ~10 minutes

✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests
  • Commit unit tests in branch feat/cmtraceopen-agent-skill

Comment @coderabbitai help to get the list of available commands.

@adamgell
adamgell merged commit b25119b into main Aug 7, 2026
13 of 14 checks passed

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR introduces a repo-checked-in “agent skill” documentation set for CMTrace Open (identity + durable memory), a thin .agents/skills/cmtraceopen wrapper skill pointing at those canonical files, and a rebuilt .Clairvoyance/ staff-org charter + routing library intended to guide parallel agent work.

Changes:

  • Add soul.md (agent identity + operating rules) and memory.md (durable repo facts + checkpoints + execution order).
  • Add a CMTrace Open specialist skill wrapper at .agents/skills/cmtraceopen/SKILL.md.
  • Add .Clairvoyance/ staff charters, notes indexes, and routing/library docs (plus kickoff prompt).

Reviewed changes

Copilot reviewed 13 out of 13 changed files in this pull request and generated 5 comments.

Show a summary per file
File Description
soul.md Adds the specialist agent “soul” (identity, non-negotiable rules, model tiering, checkpoints).
memory.md Adds “durable memory” facts (architecture summary, checkpoints, recovery branches, execution order, quick-reference paths).
library.md Updates root workspace library routing with links to the new soul/memory files.
.Clairvoyance/staff/ceo-charter.md Defines CEO charter (scope, quality bar, architecture boundary, truth requirements).
.Clairvoyance/staff/coder-charter.md Defines implementation lane charter (red-first, gates, purity rules).
.Clairvoyance/staff/ui-design-charter.md Defines UI/design charter (stable contracts only, coverage states as first-class UI).
.Clairvoyance/staff/tech-writer-charter.md Defines technical-writer charter (docs grounded in merged code + corpus).
.Clairvoyance/staff/roger/index.md Adds Roger’s routing index for SCCM notes.
.Clairvoyance/staff/theo/index.md Adds Theo’s routing index for docs-audit notes.
.Clairvoyance/memory/index.md Adds shared workspace memory index.
.Clairvoyance/library.md Adds Clairvoyance routing library for staff (where to read what).
.Clairvoyance/kickoff-prompt.md Adds a kickoff prompt for starting a CEO staffing/assessment cycle.
.agents/skills/cmtraceopen/SKILL.md Adds the “cmtraceopen” skill wrapper pointing to canonical repo files.
Suppressed comments (3)

.Clairvoyance/library.md:39

  • These paths use .clairvoyance/... but the checked-in directory is .Clairvoyance/.... Using the correct casing avoids broken links on case-sensitive filesystems / GitHub paths.
- Shared memory index: `.clairvoyance/memory/index.md`
- Staff memory lives under `.clairvoyance/staff/{name}/index.md`

.agents/skills/cmtraceopen/SKILL.md:41

  • Quick Start references repo/cmtraceopen/... paths that don't exist in this repo; using repo-root paths makes this section actionable when read from GitHub.
1. **Read `repo/cmtraceopen/soul.md`** — identity, rules, model tiers, decision framework
2. **Read `repo/cmtraceopen/memory.md`** — checkpoints, recovery branches, execution order, ecosystem state
3. **Read `~/.hermes/cmtrace-pm-charter.md`** if doing execution-manager work — has SHAs, gates, reporting style
4. **Act** — but verify against Adam's known rules before touching code

.agents/skills/cmtraceopen/SKILL.md:74

  • These Key Paths entries reference repo/cmtraceopen/..., which isn't a repo directory. Switching to actual repo-relative paths avoids confusion and broken links.
| Agent soul | `repo/cmtraceopen/soul.md` |
| Agent memory | `repo/cmtraceopen/memory.md` |
| PM charter | `~/.hermes/cmtrace-pm-charter.md` |
| Scaffold pipeline | `~/.hermes/skills/software-development/cmtrace-scaffold-pipeline/` |
| Repo routing index | `repo/cmtraceopen/library.md` |
| Agent rules | `repo/cmtraceopen/AGENTS.md` |
| Build commands | `repo/cmtraceopen/CLAUDE.md` |
| Staff org | `repo/cmtraceopen/.Clairvoyance/staff/` |

Comment thread memory.md
| Staff org charters | `.Clairvoyance/staff/` |
| Scaffold pipeline skill | `~/.hermes/skills/software-development/cmtrace-scaffold-pipeline/` |
| Execution plans | `docs/superpowers/plans/2026-07-30-sccm-*.md` (7 docs) |
| Specs | `.Clairvoyance/specs/2026-*/` |
Comment thread .Clairvoyance/library.md
# Workspace Knowledge Library

This is a routing index for Clairvoyance Staff, not a reading list.
WikiLinks in this file resolve to documents under `.clairvoyance/Docs/` or the repo root.

## When to read what

WikiLinks in this index resolve within `.clairvoyance/memory/`. Scan the IF lines, follow exactly one matching note, and skip everything else.
Comment thread memory.md
Comment on lines +92 to +95
| **CEO** | `staff/ceo-charter.md` | Reasoning (gpt-5.6-sol, claude-opus-4-8) | Runs the org; Adam runs CEO. Owns execution board (#317), quality bar, architecture boundary, budget, truth-telling |
| **Coder** | `staff/coder-charter.md` | Scaffold/Mid (kimi-k3/k2.7-code/grok-4-20-reasoning) | Implementation pool — one per issue lane. Red-first, anchor-grounded, worktree discipline, full gates |
| **UI/Design** | `staff/ui-design-charter.md` | Mid (kimi-k3) | Product designer frontend engineer — stable contracts only, coverage states as first-class UI |
| **Tech Writer** | `staff/tech-writer Charter.md` | Scaffold (kimi-k2.7-code) | Docs from merged code only — no unshipped behavior documented |
Comment on lines +22 to +26
| File | Location | Purpose |
|------|----------|---------|
| **soul.md** | `repo/cmtraceopen/soul.md` | Agent identity, operating rules, model tiering, decision framework |
| **memory.md** | `repo/cmtraceopen/memory.md` | Durable facts: architecture, verified checkpoints, recovery branches, execution order, worktree ecosystem |
| **PM charter** | `~/.hermes/cmtrace-pm-charter.md` | Execution manager contract with SHAs and per-slice gates |
adamgell added a commit that referenced this pull request Aug 13, 2026
…under Unreleased (#530)

Catch up the Unreleased section with all ten commits merged since the
last changelog update (54d539e / #513):

- Microsoft Store app evidence lane (#358 / #518)
- Reducer Framework v1 governance, ADRs, and charters (#519)
- Windows Autopilot evidence parser outside ESP (#362 / #450)
- Company Portal Windows LocalState logs (#366 / #460)
- Bounded advanced SCCM server capture (#500), folded into the
  existing native SCCM diagnostics path bullet
- Agent tooling / Clairvoyance staff org scaffolding (#516)
- Dependency bumps (quick-xml, time, install-action) and the
  GitHub Sponsors funding link

No version bump: package.json/Cargo.toml/tauri.conf.json remain at
1.5.1 with no new tag, so this stays purely an Unreleased catch-up.


Claude-Session: https://claude.ai/code/session_01A8z5Ysfa5Afts6gVPmHQZj

Co-authored-by: Claude <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

enhancement New feature or request feature New feature

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants