Skip to content

dvy1987/agent-loom / .agents/skills/quickstart

quickstart

Verified

Guided first-run that produces a real verified win in under five minutes using the skill library on a seeded offline fixture. Load when a new user asks how to start, run the demo, try agent-loom, or get a quick win. Also triggers on "quickstart", "first run", "demo agent-loom", "try the skills", or onboarding to the library. Zero external credentials required. Idempotent — safe to run multiple times.

Quickstart

You get a new user to a real, verified result in ~5 minutes — not a staged fake. Demo uses safe-change on examples/seed/calc/.

Hard Rules

Zero external credentials — offline fixture only. Invoke a real skill (safe-change) — never simulate success. Idempotent — if fix already applied, run impact map + tests and report "already done." End with one-line "what just happened" + link to examples/ tours. Total flow ≤5 minutes wall clock.


Workflow

Step 1 — Confirm install

User has agent-loom skills visible (.agents/skills/ or global install). If missing, point to README.md Installation.

Step 2 — Run demo (read references/DEMO-FLOW.md)

  1. cd examples/seed/calc (or pass path to verify.sh)
  2. dependency-mapping on divide in calc.py
  3. safe-change — add zero guard; verify fails before fix, passes after
  4. bash .agents/skills/safe-change/scripts/verify.sh examples/seed/calc

Step 3 — Explain

**What just happened:** You ran impact mapping → one verified edit → tests green.
That's the safe-change loop — the library's highest-leverage coding skill.

Step 4 — Next steps

Point to:

  • examples/safe-change-demo/README.md
  • examples/structured-planning-demo/README.md
  • examples/deploy-anywhere-demo/README.md
  • docs/why-agent-loom.md

Gotchas

  • pytest may need pip install pytest once — document if missing.
  • User on Windows: use same paths with forward slashes in agent prompts.

Output Format

## Quickstart complete

Result: [KEPT | already fixed]
Verify: pytest → [pass]
What happened: [one line]
Next: [examples link]

Examples

Full walkthrough: references/DEMO-FLOW.md


Common Rationalizations

ExcuseReality
"Fake a passing test"Destroys trust — run real safe-change.
"Skip install check"User without skills gets a confusing failure.
"Long architecture tour first"Win first, depth second.
"Use a remote repo"Needs network + creds — seed is offline.
"Combine all five skills"One win — safe-change only.

Verification

  • Real skill invoked (not simulated)
  • pytest run on seed fixture
  • User told what happened in one line
  • Next-step links provided

Red Flags

  • Simulated verify pass
  • Demo requires API keys
  • No reference to examples/

Prune Log

Last pruned: 2026-07-05

  • Initial release from high-leverage skill spec (Skill 5 family)

Impact Report

Quickstart: [pass|already-done] | Skill: safe-change | Fixture: examples/seed/calc
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 466kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 466kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 319kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 319kUpdated Claude CodeCodexVerifiedInstall