Skip to content

README.md

Latest commit

 

History

History
73 lines (47 loc) · 2.85 KB

README.md

File metadata and controls

73 lines (47 loc) · 2.85 KB

OpenWar example briefs

Seven reference briefs you can run end-to-end against any configured adapter.

creative-brief.md

A copywriting task with gated execution. Demonstrates frontmatter with scope_locked: false, mode: gated, and a soft generation_credits pre-approval.

npx @pythonluvr/openwar run examples/creative-brief.md --adapter anthropic

engineering-brief.md

A code task with auto-pilot execution and a locked scope. Demonstrates scope_locked: true, mode: auto, and filesystem_write pre-approved.

npx @pythonluvr/openwar run examples/engineering-brief.md --adapter openai --model gpt-4o

file-editing-brief.md (v0.3+, requires tools)

A real refactor: rename a symbol across every .ts file in the workdir. Demonstrates read_file, list_dir, and apply_patch native tools with filesystem_write pre-approved.

npx @pythonluvr/openwar run examples/file-editing-brief.md --adapter anthropic --workdir ./your-project

research-brief.md (v0.3+, requires tools + network)

Fetch three GitHub README files and save them locally. Demonstrates http_fetch and write_file. If you have ~/.openwar/http-allow.json configured, allow *.githubusercontent.com first.

npx @pythonluvr/openwar run examples/research-brief.md --adapter anthropic

mcp-brief.md (v0.3+, requires MCP)

Survey the workdir through the official Filesystem MCP server. Demonstrates mcp_servers in frontmatter and the mcp_tool:filesystem:* wildcard.

npx @pythonluvr/openwar run examples/mcp-brief.md --adapter anthropic

multi-agent-brief.md (v0.4+, multi-agent)

A small static-site generator built with the three default roles (planner, executor, reviewer). The planner decomposes the brief into linear sub-tasks, the executor runs each, and the reviewer evaluates against each sub-task's acceptance criteria. Demonstrates roles: and budgets: frontmatter.

# Dry-run the plan first (no execution, no spending).
npx @pythonluvr/openwar plan examples/multi-agent-brief.md --adapter anthropic

# Full run.
npx @pythonluvr/openwar run examples/multi-agent-brief.md --adapter anthropic

critic-disagreement-brief.md (v0.4+, four-role)

A tiny leap-year function with all four roles enabled (planner, executor, reviewer, critic). When the critic disagrees with the reviewer the coordinator halts on Phase 2 and asks the operator to pick a verdict.

npx @pythonluvr/openwar run examples/critic-disagreement-brief.md --adapter anthropic

Inspecting after the run

openwar list                              # show recent sessions
openwar inspect <brief_id>                # show session metadata
openwar inspect <brief_id> --transcript   # full back-and-forth

Sessions persist to ~/.openwar/sessions/. Resume with openwar resume <brief_id> (multi-agent or single-agent; the runtime detects which).