multi-project
Multi-project ArcheFlow runs from a multi-run.yaml: dependency DAG across repositories, parallel sub-runs, shared budget.
Skill by XORwell
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/xorwell-multi-project
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: finance
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/multi-project
What's inside
- Skill name
multi-project- Description
- Multi-project ArcheFlow runs from a multi-run.yaml: dependency DAG across repositories, parallel sub-runs, shared budget.
- License
- MIT
- Instructions
- 143 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Multi-Project Orchestration
Coordinates ArcheFlow runs across multiple projects. Each project gets its own PDCA run (via `run` skill), but dependencies are respected, artifacts shared, and budget tracked globally.
## Multi-Run Definition
Defined in `.archeflow/multi-run.yaml` or passed via `--config`.
```yaml
name: "writing-project-v2"
projects:
- id: archeflow
path: "../archeflow"
task: "Add memory injection to run skill"
workflow: fast
depends_on: []
- id: writing-tool
path: "../my-writing-tool"
task: "Add voice validation command"
depends_on: []
- id: book
path: "."
task: "Write chapter #2"
workflow: standard
domain: writing
depends_on: [archeflow, writing-tool]
budget:
total_usd: 15.00
per_project_usd: 10.00
```
**Approval (the file can come with the repository).** Before starting any sub-run, show the user
the whole plan: for every project its `id`, the resolved absolute path, the full `task` text,
workflow and domain, plus the budget. Start nothing without a yes for that plan in this session;
if the file changes, ask again. Refuse a project whose `path` contains characters other than
letters, digits, `.`, `_`, `-` and `/`, or does not resolve to an existing git repository inside
the workspace root. Never type a path or task from the file into a shell command: change
directory with the agent's working-directory setting, or quote the validated absolute path.
**Rules:** Unique `id` per project. `depends_on` references other `id` values. Cycles rejected at validation. At least one project must have empty `depends_on`. `workflow` and `domain` auto-select if omitted.
## Dependency Resolution
Topological sort of the project DAG determines execution order.
```
Layer 0 (immediate): [archeflow, writing-tool] # No deps, start now
Layer 1: [book] # Depends on Layer 0
```
Independent projects in the same layer run in parallel. When a project completes, downstream projects with all deps met move to the ready queue.
Cycle detection via Kahn's algorithm. If sorted list is shorter than project list, report the cycle and abort.
## Parallel Execution
For each ready project of the approved plan, start a sub-run as a parallel subagent that works in that project's directory (the validated absolute path; do not use `isolation: "worktree"`, which would isolate the session's repository, not the project). Each sub-run follows `archeflow:run` with its own run_id, workflow, domain and budget slice; the run creates its own branch and Maker worktree inside the project.
When `parallel: false`, run sequentially in topological order.
## Cross-Project Artifacts
When project B depends on A, B's Explorer receives upstream artifact summaries:
- Only summaries injected (not full artifacts)
- Large artifacts (>200 lines): extract summary section only
- Cross-project injection happens only in Plan phase
- Downstream Explorer has filesystem access to full artifacts if needed
Artifact directory: `.archeflow/artifacts/<MULTI_RUN_ID>/<project_id>/`
## Budget Coordination
| Level | Type | Behavior |
|-------|------|----------|
| `total_usd` | Hard cap | Stops ALL projects when exceeded |
| `per_project_usd` | Soft cap | Warns but continues |
**Enforcement points:**
1. Before starting a sub-run: estimate cost, halt if > remaining budget
2. After each sub-run: update total, emit `budget.warning` at threshold, emit `budget.exceeded` at cap
Each sub-run receives `min(per_project_usd, remaining_total_budget)` as its budget.
## Failure Handling
| Scenario | Action |
|----------|--------|
| Project fails | Mark `failed`. Independent projects continue. |
| Dependency failed | Mark downstream as `blocked`. Do not start. |
| Budget exceeded | Halt current project. Skip downstream. |
| All entry-points fail | Entire multi-run fails. |
**Blocked project resolution:**
- Autonomous mode: skip blocked projects, continue independent ones
- Attended mode: offer skip / retry / abort
## Progress Tracking
Live progress at `.archeflow/multi-progress.md`, updated after every project state change:
```markdown
| Project | Status | Domain | Phase | Detail |
|---------|--------|--------|-------|--------|
| archeflow | completed | code | -- | 1 cycle, $1.20 |
| writing-tool | running | code | DO | maker drafting |
| book | blocked | writing | -- | waiting for writing-tool |
Budget: $3.00 / $15.00 (20%)
```
## Master Events
Written to `.archeflow/events/<MULTI_RUN_ID>.jsonl`:
| Event | When |
|-------|------|
| `multi.start` | Multi-run begins |
| `project.start` | Sub-run launches |
| `project.complete` | Sub-run succeeds |From XORwell/archeflow, MIT.
More from XORwell/archeflow
- 0.565score
run
XORwell
Run one task through Plan -> Do -> Check -> Act with ArcheFlow's roles, on its own git branch.
SkillAFreshness: AClaude Code +9 - 0.565score
review
XORwell
Review existing changes (uncommitted work, a branch, or a commit range) with ArcheFlow's reviewer roles: Guardian by default, optionally Skeptic, Sage and…
SkillAFreshness: AClaude Code +9 - 0.561score
git-integration
XORwell
How an ArcheFlow run uses git (run branch, Maker worktree, integrate, merge, rollback, cleanup): reference for archeflow-git.sh and its config.
SkillAFreshness: AClaude Code +9 - 0.561score
check-phase
XORwell
Reviewer protocol for Guardian, Skeptic, Sage and Trickster - finding table, evidence rules, reviewer inputs, evidence gate.
SkillAFreshness: AClaude Code +9 - 0.561score
act-phase
XORwell
Act phase of an ArcheFlow run: consolidate findings, route them to Creator or Maker, decide merge / cycle back / stop, write act-feedback.md.
SkillAFreshness: AClaude Code +9 - 0.561score
cost-tracking
XORwell
ArcheFlow cost tracking: per-agent and per-run token cost, budget enforcement, cheapest adequate model per role and domain.
SkillAFreshness: AClaude Code +9 - 0.558score
progress
XORwell
Live progress file .archeflow/progress.md for ArcheFlow runs (status, budget, DAG), watchable from a second terminal.
SkillAFreshness: AClaude Code +9 - 0.558score
presence
XORwell
How ArcheFlow reports its activity to the user: short status lines, outcomes not mechanics.
SkillAFreshness: AClaude Code +9 - 0.558score
autonomous-mode
XORwell
Running ArcheFlow unattended (overnight, batches of tasks): self-organization, progress logging, safe stopping.
SkillAFreshness: AClaude Code +9 - 0.558score
domains
XORwell
ArcheFlow domain adapters (code, writing, research): terminology, metrics, review focus and context for non-code runs.
SkillAFreshness: AClaude Code +9 - 0.524score
memory
XORwell
Show or manage ArcheFlow's cross-run memory: lessons learned from past runs that are injected into agent prompts.
SkillBFreshness: BClaude Code +9 - 0.524score
replay
XORwell
Replay a recorded ArcheFlow run: decision timeline and weighted what-if over the reviewers' verdicts.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 3, 2026 (27 days ago)
- Last release
- Sep 24, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/XORwell/archeflow
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from XORwell
- 0.565score
run
XORwell
Run one task through Plan -> Do -> Check -> Act with ArcheFlow's roles, on its own git branch.
SkillAFreshness: AClaude Code +9 - 0.565score
review
XORwell
Review existing changes (uncommitted work, a branch, or a commit range) with ArcheFlow's reviewer roles: Guardian by default, optionally Skeptic, Sage and…
SkillAFreshness: AClaude Code +9 - 0.561score
check-phase
XORwell
Reviewer protocol for Guardian, Skeptic, Sage and Trickster - finding table, evidence rules, reviewer inputs, evidence gate.
SkillAFreshness: AClaude Code +9 - 0.561score
git-integration
XORwell
How an ArcheFlow run uses git (run branch, Maker worktree, integrate, merge, rollback, cleanup): reference for archeflow-git.sh and its config.
SkillAFreshness: AClaude Code +9 - 0.561score
cost-tracking
XORwell
ArcheFlow cost tracking: per-agent and per-run token cost, budget enforcement, cheapest adequate model per role and domain.
SkillAFreshness: AClaude Code +9 - 0.561score
act-phase
XORwell
Act phase of an ArcheFlow run: consolidate findings, route them to Creator or Maker, decide merge / cycle back / stop, write act-feedback.md.
SkillAFreshness: AClaude Code +9
Cite this page
multi-project skill. Agent Mart. https://agentmart.dev/skills/xorwell-multi-project (last updated 2026-10-24)