| name | ijfw-workflow |
| description | Use when the user says: 'build', 'create', 'plan', 'new project', 'brainstorm', 'design', 'UI', 'website', 'dashboard', 'app', 'help me build', 'launch', 'book', 'campaign', or anything project-level. Skill body decides Quick vs Deep path. |
| context | fork |
| model | sonnet |
IJFW Workflow
Two modes, same principles, same invariants.
- Quick -- 5 moves, 3-5 minutes, locked brief. For features, fixes, ideas.
- Deep -- 6 required modules + 3 optional, 20-45 minutes. For new projects, major refactors, launches.
Donahoe Loop: BRAINSTORM -> PLAN -> EXECUTE -> VERIFY -> SHIP -> MEASURE
|<--- memory recall at every entry --->|
|<--- Trident cross-audit on request --->|
Detailed downstream procedure (PLAN / EXECUTE / VERIFY / SHIP, including wave dispatch, post-DONE pipeline, Phase-E cross-audit): see references/post-brainstorm-workflow.md.
Deep-mode optional modules (EXTERNAL BRIEF / ANTI-SCOPE / TRIDENT CROSS-CRITIQUE): see references/deep-mode-optional.md.
RUNTIME BOOTSTRAP AND FALLBACKS
Before the first workflow write or command invocation, inspect whether .ijfw/memory/ exists so the auto-picker and empty-state opener can use that signal accurately. Then ensure .ijfw/memory/ and .planning/ exist before writing artifacts. If the ijfw CLI is unavailable in this session, continue with markdown files and visible chat checklists, then state the exact CLI command the user can run later. Optional commands such as ijfw cross, ijfw design, ijfw recover, ijfw blackboard, ijfw team, and ijfw swarm must degrade to explicit written artifacts instead of blocking the workflow.
AUTO-PICKER (runs first, every time)
Deterministic signals, visible reasoning, no friction.
| Signal | Points towards |
|---|
| Prompt < 15 words | Quick |
| Prompt has clear verb + object | Quick |
| Vague verbs alone ("improve", "fix", "handle", "deal with") | Deep |
| "New project", "major refactor", "launch", "design" | Deep |
Project dir has no .ijfw/memory/ | Deep |
| Explicit "brainstorm", "quick idea", "just sketch" | Quick |
Protocol:
- Read signals silently.
- Say in one line:
Reading this as <Quick|Deep> -- <reason>. Say "go deeper" / "just quick" to switch.
- If signals tie, ask once:
Quick or Deep? Accept any affirmative shortcut ("q" / "d").
- Start immediately. The user should never wait on a modal.
Mid-flow escalation: user can say go deeper at any Quick step; skill re-enters Deep at the equivalent module. Mid-flow de-escalation: just quick collapses the remaining Deep modules into a single LOCK.
EMPTY-STATE OPENER
First session in a project (no .ijfw/memory/ or zero entries in it) is the onboarding moment. Do not stay silent. Open with one line:
Clean slate here. Want me to run a 5-minute Quick brainstorm on what we're building, or jump straight in?
User replies brainstorm / jump in / custom intent. If they jump in, still offer memory hooks for the first 3 turns so decisions get captured. If they brainstorm, route into QUICK mode FRAME. Either way, .ijfw/memory/ gets bootstrapped silently.
If memory is populated but the last handoff is >7 days old, open with a softer beat: Welcome back -- last handoff was <N> days ago. Quick recap? User says recap / new task / actual intent.
BRAINSTORM DISCIPLINE (invariants)
Hard rules. Violating any of these is a workflow failure worth auditing.
- One question at a time. Never dump a numbered list and wait. Ask, get the answer, absorb, ask the next.
- No offscreen research. If you dispatch an Explore / scout agent, paste a synthesis (3-5 bullets + contradictions + plan implications) in-chat BEFORE using it for anything.
- No skipping to the plan.
plan.md is written only after the user has explicitly confirmed the brief.
- No auto-advance. Audit gates are user-facing checklists, not silent passes.
- Visible deliverables. Every artifact (brief.md, research.md, plan.md) is summarized to the user in-chat when written.
- Intermediate thinking is tight output, not monologue. Thirty words, then the next question.
Failure signatures to catch in yourself: about to write plan.md without user confirming brief; about to dispatch a research agent whose output will not be paraphrased back; about to say "Phase N complete, ready to build" in a turn where the user has not seen the intermediate findings.
MEMORY HOOK (every FRAME step)
At the start of every brainstorm or plan, call ijfw_memory_recall with the goal text when the memory tool is available. If it is unavailable, read the visible memory files under .ijfw/memory/ when possible; if neither is available, continue and say clean slate -- memory unavailable this turn.
I remember: decision from on -- <1-line summary>. Pull full context?
This is the single biggest superpower IJFW delivers. Never skip the attempt. If memory is empty, say so ("clean slate -- nothing recalled") so the silence is intentional, not absence of effort.
QUICK MODE -- 5 moves, 3-5 min
For focused work. Each move has ONE input slot.
Move 1 -- FRAME (45s)
- Parse the goal from the ask or ask:
Goal in one line.
- Memory hook fires. Paste up to 3 related recalls inline.
- Rewrite vague asks into verifiable goals before echoing back. Examples: "Add validation" → "Tests for invalid inputs, then make them pass"; "Fix the bug" → "Failing repro test, then make it pass"; "Refactor X" → "Suite passes before and after, no public API changes"; "Make it faster" → "Profile hot path, change it, show the bench improved"; "Clean up the code" → "Pick one smell, fix only that, one-commit diff".
- If the ask cannot be reduced to a checkable outcome, surface that gap before proceeding.
- Echo:
So: <concise goal>. Yes? User confirms or edits.
Move 2 -- WHY (30s)
Ask: Why does this matter? What's broken if we don't ship it? One 5-Whys drill if surface. Surface the root: Root: <X>. That means we should <design implication>.
Move 3 -- SHAPE (60s)
Propose 3 approaches, each 1 line + 1 tradeoff. User picks, hybrids, or overrides. No blank page.
Move 4 -- STRESS (30s)
Pre-mortem flash: Top risk: <concrete scenario>. Mitigation: <concrete fix>. Sutherland wow: the risk the user hadn't thought of.
Move 5 -- LOCK (15s)
- Paste the brief in-chat (max 6 lines: goal / root / approach / risk / mitigation / success).
- One word:
lock / fix <X> / go deeper.
- On
lock: write .ijfw/memory/brief.md. Route straight to PLAN (see references/post-brainstorm-workflow.md).
- Invoke
ijfw-agents-md skill with intent context (brief.md) to seed AGENTS.md if missing; create platform adapter (e.g. CLAUDE.md) from template if detected IDE file is absent.
- N7 LOCK-hook: call
populateDisciplineBlock(projectRoot) (from mcp-server/src/orchestrator/agents-md-blackboard.js) immediately after ijfw-agents-md returns. This writes the DISCIPLINE block to AGENTS.md with the project-type-appropriate discipline template. Non-throwing -- a failure here is logged but never blocks the workflow. Idempotent re-fire at plan-LOCK; byte-identical output unless .ijfw/memory/brief.md frontmatter type: was manually changed between locks.
Quick-mode closer: You went from <original-ask> to locked brief with <N> risks mitigated in <M> minutes.
DEEP MODE -- 6 required modules + 3 optional
Modules are a spine, not a checklist. Every module has a memory hook, a visible artifact, and a one-word commit.
Module 1 -- FRAME (3 min)
Memory recall on goal keywords. Socratic arc: problem → users → constraints → scope. One question per turn. Echo back every 3 turns. Artifact: .ijfw/memory/brief-draft.md (≤30 lines); promote to brief.md only after LOCK.
Module 2 -- RECON (5-10 min)
State research questions in-chat first. Dispatch scout / Explore agents (parallel where independent), or do passes locally and record the limitation. Paste synthesis: ask + answer + contradictions + plan implications. Artifact: .ijfw/memory/research.md.
Module 3 -- HMW (3 min)
Propose 2-3 "How Might We" reframings based on FRAME + RECON. User picks, rejects, or edits. Chosen HMW anchors DIVERGE.
Module 4 -- DIVERGE (8 min)
Sketch 4-5 approaches as 2-line bullets (shape + tradeoff). User picks 2, rejects, or hybrid A + C. Generate a 6th on demand.
Module 5 -- CONVERGE (5 min)
Draft success metrics / acceptance criteria. Pre-mortem: 4-5 plausible failures. User picks top 2 risks; propose mitigations. Append metrics + risks + mitigations to brief.md.
Module 6 -- LOCK (2 min)
Paste the full brief (goal / HMW / approach / metrics / risks / mitigations). Optional Trident cross-critique fires here if ENABLED. User says lock / fix <X> / skip Trident / route to plan. On lock: promote brief-draft.md → brief.md, route to PLAN, invoke ijfw-agents-md skill. N7 LOCK-hook fires (see N7 description above).
Optional modules: EXTERNAL BRIEF, ANTI-SCOPE, TRIDENT CROSS-CRITIQUE — see references/deep-mode-optional.md for triggers + procedure.
POST-BRAINSTORM WORKFLOW (overview)
After LOCK, the brief drives every downstream phase. Same discipline, same memory hooks, same positive framing. Full procedure: references/post-brainstorm-workflow.md.
Quick reference:
- PLAN → write
.ijfw/memory/plan.md; design auto-fire on visual artifacts; ijfw plan-check audit.
- EXECUTE → wave dispatch via
dispatch-planner.js; status-block protocol; mandatory ijfw_state MCP call with verb: 'subagent.post-done' after every subagent (v1.5.0 T13 — single state-SDK MCP face); task micro-audit.
- VERIFY → audit against brief, not plan; Functional + UX + Security + Quality checklists; optional
ijfw cross audit <diff>.
- Cross-Audit Phase (Phase E) → auto-fires after VERIFY; reads
.ijfw/swarm.json::auditors or falls back to [codex, gemini, claude]; PASS / CONDITIONAL / FAIL routes.
- SHIP → atomic commit, explicit user approval before tag/release/publish, memory write, ship gate.
NARRATION (every transition)
One sentence at every phase entry and mid-step ping. Format: Phase <name> -- Move <n> -- <what's happening>.
Model routing (mandatory): Before every agent dispatch, name the actual platform/model or role tier available, e.g. Routing to builder for implementation or Routing to Sonnet for this build. Never dispatch silently. No hardcoded phase numbers — narration tracks the current workflow step.
POSITIVE FRAMING (enforced everywhere)
Replace negatives with reframes for brainstorming, planning, and ordinary user-facing progress. Do NOT rewrite exact failure terminology in audit, CI, preflight, security, exception, test, or log contexts.
| Never | Always |
|---|
| "found problems" | "surfaced X points" |
| "failed" | "didn't complete -- try again?" |
| "error" (as header) | "heads up" / "one thing" |
| "missing" | "ready to add" |
| "not supported" | "standing by" |
| "broken" | "needs a sharpening pass" |
End-of-phase closer is a receipt: You went from <input> to <outcome> in <time>.
USER OVERRIDES (one-word commits)
lock / go deeper / just quick / skip <module> / force Trident / rollback / help.
TASK TRACKING USAGE (mandatory)
When the platform exposes native task tracking, create one task per specialist or prepared swarm task before dispatch. Mark in_progress when work starts, then completed or blocked as soon as each worker reports back. If no native tracker exists, use .ijfw/blackboard/tasks.json plus concise progress updates.
[Model] prefix required -- every task title includes the model tier: [Haiku] Scout: ..., [Sonnet] Build: ..., [Opus] Audit: ....
Quick-mode minimum: 5 tasks (one per move). Deep-mode minimum: one per module + one per specialist + one per audit gate + ship gate. Silent dispatch is a workflow violation.
NAMING-GAP AUDIT (every turn)
Before emitting any "next step" text, scan for foreign plugin prefixes -- any <plugin>: pattern where <plugin> is not ijfw. If found as an action verb, rewrite to the IJFW-native equivalent or halt with: Rewrite needed -- foreign plugin verb detected.
Specialist swarm members (code-reviewer, silent-failure-hunter, pr-test-analyzer, type-design-analyzer) are allowed. Foreign plugin commands are not.
STATE FILE (Deep Mode)
Write .ijfw/state/workflow.json at every transition with {mode, module, last_commit, artifacts, next}. On session resume: read this file, echo the current state, offer continue / restart. Memory recall also fires on resume.
Invariant: every move should make the user feel smarter and more in control -- memory surfaces forgotten context, Assistant proposes before the user has to, Trident challenges before they commit, one word advances. Anything that makes them feel stupid or stuck is a workflow bug.
