@osfactory/har 1.0.0 → 1.2.0

package/dist/templates/AGENTS.md.template

@@ -20,7 +20,7 @@ harness or report it; do not fall back to ad-hoc commands.
 
 Occupied slots always block: `complete` / `teardown`, then launch. Customize the
 harness only through `harness.env`, `stages.json` + `.har/stages/`, `.har/hooks/`
-and `.har/plugins/` — the generated `.har/*.sh` shims are not an editing surface.
+and `.har/plugins/` — there is no generated `.har/*.sh` entry-point surface.
 
 Full detail — slot environment, readiness, definition of done, project commands,
 commit gate: [`.har/README.md`](.har/README.md) and [`.har/stages.json`](.har/stages.json).

package/dist/templates/adaptation-prompt-init.md

@@ -5,8 +5,8 @@ Adapt the `.har/` harness in this repository so AI coding agents can run the pro
 Explore this repository, then edit files in `.har/` directly to make the harness runnable for this project.
 
 `.har/` is a **configuration surface** — the runtime machinery lives in the HAR
-package behind the `./.har/*.sh` shims. Never edit the shims or invent new config
-files; every customization has a sanctioned home:
+package. CLI and MCP are the only entry points. Never add lifecycle wrappers
+or invent new config files; every customization has a sanctioned home:
 
 | Customization | Home |
 |---------------|------|
@@ -16,9 +16,9 @@ files; every customization has a sanctioned home:
 | Anything bigger | Plugins (`har env add-plugin`, `har plugin create` for local) |
 | Full runtime ownership (rare) | `har env eject` |
 
-## HAR profiles (pick the right one at init)
+## HAR profiles (pick the right one at scaffold)
 
-`har env init` scaffolds from one of three boilerplate profiles — **choose the profile that matches this repository**:
+The scaffold comes from one of three boilerplate profiles — **choose the profile that matches this repository**:
 
 | Profile | Best for | What you get |
 |---------|----------|--------------|
@@ -26,11 +26,7 @@ files; every customization has a sanctioned home:
 | `cli` | **CLI tools, libraries, test-suite repos** (typical SWE-bench) | Git worktree by default, no PM2; optional Docker for databases; run project commands in isolation |
 | `ios` | **Mobile iOS / Swift** | xcodebuild + iOS Simulator; scheme/project/simulator in `harness.env` |
 
-```bash
-har env init                  # default (SaaS/web)
-har env init --profile cli    # libraries / CLI / polyglot test repos
-har env init --profile ios    # iOS apps
-```
+If you must re-scaffold (destructive): `har env init --force --profile default|cli|ios`.
 
 Do **not** disable worktrees, rewrite launch into repo-root-only mode, or pick the wrong profile unless the project truly requires it.
 
@@ -55,8 +51,8 @@ Replace all TODO placeholders. Key files:
 Identify the **primary application** — the ONE app coding agents will modify and run per-slot. Everything else is shared and runs once for all slots:
 
 1. **Primary app** → set `HARNESS_PRIMARY_APP` in `harness.env`; wire ONLY its dev processes into `ecosystem.agent.template.cjs` (a primary app may still need several processes, e.g. api + frontend of the same app).
-2. **External dependencies** (database, cache, queue, mail, ...) → keep/add them as services in `docker-compose.agent.yml`, delete the menu entries the project doesn't use, and list the needed ones in `HARNESS_INFRA_SERVICES` (e.g. `"db redis"`). They start once via `setup-infra.sh` on fixed ports and serve every slot.
-3. **Internal supporting services** (a monolith/monorepo's other services the agent depends on but is not changing) → do NOT start them per-slot. Run them once and shared: as compose services in `docker-compose.agent.yml`, or as PM2 processes in `.har/ecosystem.shared.config.cjs` (processes named `har-shared-<name>`, started automatically by `setup-infra.sh` when the file exists). Point the primary app at them through `env.template`.
+2. **External dependencies** (database, cache, queue, mail, ...) → keep/add them as services in `docker-compose.agent.yml`, delete the menu entries the project doesn't use, and list the needed ones in `HARNESS_INFRA_SERVICES` (e.g. `"db redis"`). They start once via `har env setup-infra` on fixed ports and serve every slot.
+3. **Internal supporting services** (a monolith/monorepo's other services the agent depends on but is not changing) → do NOT start them per-slot. Run them once and shared: as compose services in `docker-compose.agent.yml`, or as PM2 processes in `.har/ecosystem.shared.config.cjs` (processes named `har-shared-<name>`, started automatically by `har env setup-infra` when the file exists). Point the primary app at them through `env.template`.
 
 Simple single-app repos need none of the extra machinery: one primary app, usually one `db` in `HARNESS_INFRA_SERVICES`, no shared ecosystem file.
 
@@ -76,8 +72,8 @@ Verification is data: `.har/stages.json` lists the pipeline in
 `verificationStages` (execution order), and every id resolves to a registered
 stage with `"tier": "quick" | "full"`. Adapt verification by editing the stage
 entries — commands, tiers, order — or adding stage scripts (see
-`.har/STAGES.md`). **Never edit `.har/verify.sh`**; it is a thin iterator that
-runs the registry. The scaffolded defaults (`typecheck`, `unit-tests`, `lint`,
+`.har/STAGES.md`). Do not add lifecycle wrappers — `har env verify` runs the
+registry. The scaffolded defaults (`typecheck`, `unit-tests`, `lint`,
 `readiness`, plus `api-health` on web profiles) are derived from
 `HARNESS_ECOSYSTEM` — a starting point, not the repo's final contract. Replace
 conventions that do not match this project. Use toolchain variables from
@@ -166,8 +162,8 @@ The harness reference agents read: slot environment, readiness / what "agent
 usable" means here, definition of done, project commands, architecture.
 
 ### `.har/env.template`, `docker-compose.agent.yml`
-Adapt as needed for the project's infra (`setup-infra.sh` is a shim — the
-services it starts come from these two files plus `HARNESS_INFRA_SERVICES`).
+Adapt as needed for the project's infra (`har env setup-infra` starts
+services from these two files plus `HARNESS_INFRA_SERVICES`).
 
 ### Port allocation & shared services
 
@@ -211,15 +207,14 @@ Coding agents discover the harness through two files:
 
 Do **not** create `AGENT.md` (singular) — that legacy name is obsolete. If `AGENT.md` exists, merge any unique notes into `AGENTS.md` and delete `AGENT.md`.
 
-If **`har env init` already created `AGENTS.md`**, refresh the **HAR / agent environment** section and fill in **Project-specific notes** — do not wipe unrelated content.
+If **the scaffold already created `AGENTS.md`**, refresh the **HAR / agent environment** section and fill in **Project-specific notes** — do not wipe unrelated content.
 
 If **no `AGENTS.md` exists**, create one at the repo root using this structure:
 
 - Link to `.har/README.md`
 - State plainly: **the harness is how you run this project** — to see the app live (manual testing, browser, screenshots), `launch` a slot; never hand-roll docker/dev-server startup, and never work around a failing harness command with ad-hoc setup (fix or report it instead)
 - Commands: HAR MCP tools or `har env launch/verify/teardown`
-- Shell shims: `./.har/launch.sh`, `./.har/verify.sh`, `./.har/teardown.sh` — same runtime and run records
-- Rules (no hardcoded ports, use `./.har/agent-cli.sh`, do not touch other agents' resources)
+- Rules (no hardcoded ports, use `har env agent <id> …`, do not touch other agents' resources)
 - Project-specific notes (stack, credentials, definition of done)
 
 If **`AGENTS.md` already exists** (project-owned), add or update a concise **HAR / agent environment** section — do not replace unrelated content.
@@ -236,14 +231,14 @@ If this repository contains **more than one project or `.har` harness** (check f
 
 Include a **Run history** subsection:
 
-- Every entry point (`./.har/*.sh` shims, `har env …`, MCP) writes the same run records to `.har/runs/YYYY-MM-DD/HH-mm-ss_<stageId>_agent-<id>.json`
+- Every entry point (`har env …`, MCP) writes the same run records to `.har/runs/YYYY-MM-DD/HH-mm-ss_<stageId>_agent-<id>.json`
 - With worktrees, code runs in the worktree but run JSON lives in the main checkout `.har/runs/`
-- The `./.har/*.sh` shims forward to the same runtime — evidence records are identical from any surface
+- CLI and MCP are the only entry points — evidence records are identical from either surface
 
 ## Step 4 — Adaptation completeness (required)
 
 Before finishing, confirm the harness describes this project truthfully — do
-**not** go hunting for machinery to prune (the shims and packaged runtime are
+**not** go hunting for machinery to prune (the packaged runtime is
 not yours to trim, and wholesale deletion just manufactures drift):
 
 - [ ] No `TODO` placeholders remain anywhere in `.har/`
@@ -256,10 +251,10 @@ not yours to trim, and wholesale deletion just manufactures drift):
 
 ## Rules
 
-1. Customize only through the contract: config in `harness.env`, stages in `stages.json` / `.har/stages/`, hooks in `.har/hooks/`, plugins in `.har/plugins/` — never edit the generated `*.sh` shims
+1. Customize only through the contract: config in `harness.env`, stages in `stages.json` / `.har/stages/`, hooks in `.har/hooks/`, plugins in `.har/plugins/` — never add lifecycle `.sh` wrappers
 2. Always update `.har/README.md` to reflect current harness state
 3. Reuse existing project commands from package.json, Makefile, CI, etc.
 4. Replace all TODO placeholders
 5. Do not edit `.har/manifest.json` — managed by the har CLI
 
-When finished, summarize what you changed, confirm verification commands (`har env verify 1 --full` or `./.har/verify.sh 1 --full`) are correct for this stack, and record the adaptation with `har env maintain --finalize --summary "<what changed>"`.
+When finished, summarize what you changed, confirm verification commands (`har env verify 1 --full`) are correct for this stack, and record the adaptation with `har env maintain --finalize --summary "<what changed>"`.

package/dist/templates/adaptation-prompt-maintain.md

@@ -81,7 +81,7 @@ scripts for a lifecycle side effect, move that code into a hook.
 Platform-shape upgrades are **code, not a checklist**: the har CLI carries a
 versioned migration registry keyed on the manifest's `runtimeVersion`.
 `har env maintain` detects an old harness shape, writes `.har/MIGRATE-PROMPT.md`,
-and `har env maintain --migrate` applies the mechanical steps (shims, pure-config
+and `har env maintain --migrate` applies the mechanical steps (delete leftover wrappers, pure-config
 `harness.env` with `HARNESS_INFRA_SERVICES` / `HARNESS_INFRA_PORT_LANES`
 conversions, machinery removal) with backups under `.har/migrate/backup/`.
 

package/dist/templates/adaptation-prompt-plugin.md

@@ -31,7 +31,7 @@ stage entries in `.har/stages.json`. Never edit the shims.
 ## Prove it green
 
 ```bash
-har env launch 1                # or ./.har/launch.sh 1
+har env launch 1
 har env verify 1 --full         # must pass INCLUDING {{STAGE_IDS}}
 ```
 

package/dist/templates/agent-skills/factory-line.md

@@ -0,0 +1,96 @@
+# Run a factory line
+
+Execute **one station** of a declared multi-station program:
+
+sync → wave plan → parallel HAR slots → cumulative gate → human handoff.
+
+The program lives in a **line template** (a `*.line.json` the user points at,
+or `.har/line.json` when that file exists). This workflow is the orchestrator.
+Station-specific how-to lives in other skills the template lists.
+
+HAR already has the primitives — work units, isolated slots, stages, plugins,
+validation records. A line is composition plus a manifest. Do not invent a
+second runner, a stack-specific MCP tool, or GitHub-shaped stations.
+
+This workflow is the same on Claude Code (`/factory-line`), Cursor
+(`.cursor/commands/factory-line`), and Codex (`har-factory-line`).
+
+## Inputs
+
+Resolve these before doing anything (ask only if not inferable):
+
+1. **Line file** — path to a `*.line.json`. If the repo has no line file and
+   the user did not name one, stop and say so. Do not invent GitHub issue
+   numbers or a fixture repo.
+2. **Station id** — default: the first station whose bound work is not all
+   done / whose gate has not passed.
+3. **Slot budget** — `har env status` first. One slot per concurrent agent.
+   Occupied slots always block.
+
+## Contract (minimum)
+
+A line file is JSON with `contractVersion: 1`. Required:
+
+- `id`, `stations[]` (ordered; each has `id` + `title`)
+- `gate.cumulative` must be `true`
+- `handoff.autonomousShip` must be `false`
+
+Optional: `skills[]` and `mcp[]` (declarations + install hints — HAR does not
+vendor third-party skill packs or add core MCP tools), `plugins[]`,
+`gate.stages[]` with `fromStation` (that stage stays required from this
+station **on**), `extraStages[]`, `traveler`, `stations[].work` / `waves`.
+
+`work.source` is `github`, `linear`, `none`, or a free string — the tracker
+is data on the station, not the type of the station. If `waves` is omitted,
+the station is one sequential cell.
+
+A QA station is never removed. Adding a station must not drop earlier
+`gate.stages`. When a review finds a new class of defect, add a stage or
+doctor question and tag `fromStation` — that is how the ratchet grows.
+
+## What this workflow must not bake in
+
+Stacked PRs, a long-lived integration branch, a named fixture repo, `M0..M5`,
+and "one issue = one PR = one slot" are **instance tactics**. If the template
+does not declare them, do not do them.
+
+## 0. Load and preflight
+
+1. Read the line file. Confirm `gate.cumulative` and `handoff.autonomousShip`.
+2. Skills and MCP: confirm required ones are present. Missing optional:
+   skip those steps and say so. Do not install third-party packs yourself.
+3. Print: line id, station id, waves, slot plan, and every gate stage whose
+   `fromStation` is this station or an earlier one.
+
+## 1. Wave plan
+
+Print `stations[].waves` (or a single cell). One HAR slot per group in a
+wave. Never share a slot between two concurrent groups.
+
+## 2. Execute waves
+
+For each wave, one subagent per group, in parallel. Each:
+
+1. Checks the slot is free. Occupied → complete/teardown that slot, or pick
+   a free slot in budget.
+2. `har env launch <slot>` with `--work-id` / `--work-source` / `--work-url`
+   / `--work-title` when work is bound. Skip tracker flags when source is
+   `none`.
+3. Edits **only** in the returned work dir.
+4. Full verify in-slot until green. If this station made a new gate question
+   askable, add it — never remove an earlier station's stages.
+5. Reports back. Subagents never push, PR, complete, or teardown on their own.
+
+## 3. Gate
+
+Run every gate stage whose `fromStation` ≤ current station. If
+`gate.optInEnv` is set, export it as `1` so the opt-in jig actually runs.
+
+Red gate → fix, re-run. Never hand off a red gate. Never bypass
+(`HAR_SKIP_GATE`, `--no-verify`).
+
+## 4. Handoff
+
+Update the `traveler` the template names. Then **stop and wait**. Do not
+merge, push to the default branch, complete a slot, or release. Present a
+session handoff (summary, session branch, preview URLs, gate evidence).

package/dist/templates/agent-skills/har-wt.md

@@ -16,7 +16,7 @@ har env status        # or ./.har/agent-cli.sh 1 status
 ## 2. Launch — BEFORE editing any file
 
 ```bash
-har env launch <id>   # or ./.har/launch.sh <id>
+har env launch <id>
 ```
 
 Launch creates a fresh session worktree from HEAD and prints its **work dir** (also recorded in `.har/slots/agent-<id>.json`, path like `~/worktrees/<base>-<sha4>-har-agent-<id>-<rand4>`). If a previous launch failed partway, retry with `--resume` instead of starting fresh.

package/dist/templates/agent-skills/setup-har.md

@@ -42,15 +42,15 @@ Tell the user which profile you picked and why before continuing.
 
 ## 4. Initialize the harness
 
-```bash
-har env init --profile <profile>   # omit --profile for default
-```
+Call `har_init_harness` with the chosen `profile` (`default`, `cli`, or `ios`).
+Do not run `har onboard` (that wizard is for humans) and do not advertise
+`har env init` as the first-run command.
 
 If `.har/` already exists, stop and suggest `/har-maintain` instead.
 
 ## 5. Perform the adaptation yourself
 
-`har env init` prints an adaptation prompt and writes it to `.har/ADAPT-PROMPT.md`. Read that file and **execute its instructions yourself, now, in this session** — tailor the harness configuration surface (`harness.env`, `stages.json` + `.har/stages/`, `.har/hooks/`, `docker-compose.agent.yml`, `env.template`) and `AGENTS.md` to this repository's real stack, ports, and commands. The `./.har/*.sh` files are generated shims over the packaged runtime — never edit them.
+`har_init_harness` writes an adaptation prompt to `.har/ADAPT-PROMPT.md`. Read that file and **execute its instructions yourself, now, in this session** — tailor the harness configuration surface (`harness.env`, `stages.json` + `.har/stages/`, `.har/hooks/`, `docker-compose.agent.yml`, `env.template`) and `AGENTS.md` to this repository's real stack, ports, and commands. The `./.har/*.sh` files are generated shims over the packaged runtime — never edit them.
 
 ## 6. Register the project's checks as stages
 

package/dist/templates/agent-skills/skills.manifest.json

@@ -43,6 +43,19 @@
       },
       "cursor": { "command": true },
       "codex": { "promptName": "har-maintain" }
+    },
+    {
+      "id": "factory-line",
+      "body": "factory-line.md",
+      "title": "Run one station of a factory line",
+      "claude": {
+        "frontmatter": {
+          "name": "factory-line",
+          "description": "Execute one station of a declared multi-station program (*.line.json): parallel HAR slots, cumulative gate, human handoff. Use when asked to run a factory line, the next station, or a milestone program."
+        }
+      },
+      "cursor": { "command": true },
+      "codex": { "promptName": "har-factory-line" }
     }
   ]
 }

package/dist/templates/har-boilerplate/harness.env

@@ -1,5 +1,5 @@
 # Shared harness configuration — edit to match your project
-# Generated by har env init, maintained over time with har env maintain
+# Generated when the harness is scaffolded, maintained with har env maintain
 #
 # This file is pure KEY=value config (plus comments) and is schema-validated
 # on read. Runtime helpers (har_infra_enabled, har_pg, port resolution) live

package/dist/templates/har-boilerplate/justfile

@@ -2,38 +2,37 @@
 
 set shell := ["bash", "-cu"]
 
+repo := justfile_directory() + "/.."
+
 setup:
-    ./setup-infra.sh
+    har env setup-infra --repo {{repo}}
 
 launch id:
-    ./launch.sh {{id}}
+    har env launch {{id}} --repo {{repo}}
 
 teardown id:
-    ./teardown.sh {{id}}
+    har env teardown {{id}} --repo {{repo}}
 
 verify id:
-    ./verify.sh {{id}}
+    har env verify {{id}} --repo {{repo}}
 
 verify-full id:
-    ./verify.sh {{id}} --full
+    har env verify {{id}} --full --repo {{repo}}
 
 status:
-    #!/usr/bin/env bash
-    for id in 1 2 3 4 5; do
-        ./agent-cli.sh "$id" status 2>/dev/null || true
-    done
+    har env status --repo {{repo}}
 
 launch-all count:
     #!/usr/bin/env bash
     for id in $(seq 1 {{count}}); do
-        ./launch.sh "$id"
+        har env launch "$id" --repo "{{repo}}"
     done
 
 teardown-all:
     #!/usr/bin/env bash
     for id in 1 2 3 4 5; do
-        ./teardown.sh "$id" 2>/dev/null || true
+        har env teardown "$id" --repo "{{repo}}" 2>/dev/null || true
     done
 
 cli id cmd *args:
-    ./agent-cli.sh {{id}} {{cmd}} {{args}}
+    har env agent {{id}} {{cmd}} {{args}} --repo {{repo}}

package/dist/templates/har-boilerplate/stages.json

@@ -13,14 +13,12 @@
     {
       "id": "setup-infra",
       "kind": "setup",
-      "description": "Shared infrastructure (Docker, template DB)",
-      "command": "./.har/setup-infra.sh"
+      "description": "Shared infrastructure (Docker, template DB)"
     },
     {
       "id": "launch",
       "kind": "launch",
       "description": "Launch an isolated agent environment slot",
-      "command": "./.har/launch.sh {agentId}",
       "artifacts": [
         {
           "path": ".env.agent.{agentId}",
@@ -33,7 +31,6 @@
       "id": "verify",
       "kind": "verify",
       "description": "Verification pipeline: runs stages.json verificationStages in order (tier \"quick\" by default; --full runs the whole list)",
-      "command": "./.har/verify.sh {agentId}",
       "acceptsArgs": [
         "--full"
       ],
@@ -49,7 +46,6 @@
       "id": "status",
       "kind": "inspect",
       "description": "Inspect a running agent environment slot",
-      "command": "./.har/agent-cli.sh {agentId} status",
       "artifacts": [
         {
           "path": ".har/logs",
@@ -61,14 +57,12 @@
     {
       "id": "logs",
       "kind": "inspect",
-      "description": "Read recent logs for an agent slot",
-      "command": "./.har/agent-cli.sh {agentId} logs"
+      "description": "Read recent logs for an agent slot"
     },
     {
       "id": "teardown",
       "kind": "teardown",
-      "description": "Stop processes and release agent slot resources",
-      "command": "./.har/teardown.sh {agentId}"
+      "description": "Stop processes and release agent slot resources"
     },
     {
       "id": "typecheck",

package/dist/templates/har-boilerplate-cli/harness.env

@@ -1,5 +1,5 @@
 # CLI / library harness — optional Docker via HARNESS_INFRA_SERVICES, no PM2
-# Generated by har env init --profile cli
+# Generated when the harness is scaffolded (--profile cli)
 #
 # This file is pure KEY=value config (plus comments) and is schema-validated
 # on read. Runtime helpers (har_infra_enabled, har_pg, port resolution) live

package/dist/templates/har-boilerplate-cli/justfile

@@ -2,26 +2,25 @@
 
 set shell := ["bash", "-cu"]
 
+repo := justfile_directory() + "/.."
+
 setup:
-    ./setup-infra.sh
+    har env setup-infra --repo {{repo}}
 
 launch id:
-    ./launch.sh {{id}}
+    har env launch {{id}} --repo {{repo}}
 
 teardown id:
-    ./teardown.sh {{id}}
+    har env teardown {{id}} --repo {{repo}}
 
 verify id:
-    ./verify.sh {{id}}
+    har env verify {{id}} --repo {{repo}}
 
 verify-full id:
-    ./verify.sh {{id}} --full
+    har env verify {{id}} --full --repo {{repo}}
 
 status:
-    #!/usr/bin/env bash
-    for id in 1 2 3; do
-        ./agent-cli.sh "$id" status 2>/dev/null || true
-    done
+    har env status --repo {{repo}}
 
 cli id cmd *args:
-    ./agent-cli.sh {{id}} {{cmd}} {{args}}
+    har env agent {{id}} {{cmd}} {{args}} --repo {{repo}}

package/dist/templates/har-boilerplate-cli/stages.json

@@ -13,14 +13,12 @@
     {
       "id": "setup-infra",
       "kind": "setup",
-      "description": "Optional shared Docker infrastructure",
-      "command": "./.har/setup-infra.sh"
+      "description": "Optional shared Docker infrastructure"
     },
     {
       "id": "launch",
       "kind": "launch",
       "description": "Prepare agent slot (git worktree, deps, env file)",
-      "command": "./.har/launch.sh {agentId}",
       "artifacts": [
         {
           "path": ".env.agent.{agentId}",
@@ -33,7 +31,6 @@
       "id": "verify",
       "kind": "verify",
       "description": "Verification pipeline: runs stages.json verificationStages in order (tier \"quick\" by default; --full runs the whole list)",
-      "command": "./.har/verify.sh {agentId}",
       "acceptsArgs": [
         "--full"
       ],
@@ -49,7 +46,6 @@
       "id": "status",
       "kind": "inspect",
       "description": "Inspect agent slot status and worktree path",
-      "command": "./.har/agent-cli.sh {agentId} status",
       "artifacts": [
         {
           "path": ".har/logs",
@@ -61,8 +57,7 @@
     {
       "id": "teardown",
       "kind": "teardown",
-      "description": "Remove worktree and env file for the slot",
-      "command": "./.har/teardown.sh {agentId}"
+      "description": "Remove worktree and env file for the slot"
     },
     {
       "id": "typecheck",

package/dist/templates/har-boilerplate-ios/harness.env

@@ -1,5 +1,5 @@
 # iOS mobile app harness — Xcode / iOS Simulator, no PM2, no Docker infra
-# Generated by har env init --profile ios
+# Generated when the harness is scaffolded (--profile ios)
 #
 # This file is pure KEY=value config (plus comments) and is schema-validated
 # on read. Runtime helpers (har_infra_enabled, port resolution) live in

package/dist/templates/har-boilerplate-ios/justfile

@@ -2,29 +2,28 @@
 
 set shell := ["bash", "-cu"]
 
+repo := justfile_directory() + "/.."
+
 setup:
-    ./setup-infra.sh
+    har env setup-infra --repo {{repo}}
 
 launch id:
-    ./launch.sh {{id}}
+    har env launch {{id}} --repo {{repo}}
 
 teardown id:
-    ./teardown.sh {{id}}
+    har env teardown {{id}} --repo {{repo}}
 
 verify id:
-    ./verify.sh {{id}}
+    har env verify {{id}} --repo {{repo}}
 
 verify-full id:
-    ./verify.sh {{id}} --full
+    har env verify {{id}} --full --repo {{repo}}
 
 status:
-    #!/usr/bin/env bash
-    for id in 1 2 3; do
-        ./agent-cli.sh "$id" status 2>/dev/null || true
-    done
+    har env status --repo {{repo}}
 
 cli id cmd *args:
-    ./agent-cli.sh {{id}} {{cmd}} {{args}}
+    har env agent {{id}} {{cmd}} {{args}} --repo {{repo}}
 
 simulator:
     xcrun simctl list devices available | grep -E 'iPhone|iPad' | head -20

package/dist/templates/har-boilerplate-ios/stages.json

@@ -13,14 +13,12 @@
     {
       "id": "setup-infra",
       "kind": "setup",
-      "description": "Boot iOS Simulator and start optional shared infra",
-      "command": "./.har/setup-infra.sh"
+      "description": "Boot iOS Simulator and start optional shared infra"
     },
     {
       "id": "launch",
       "kind": "launch",
       "description": "Prepare agent slot (git worktree, iOS deps, env file)",
-      "command": "./.har/launch.sh {agentId}",
       "artifacts": [
         {
           "path": ".env.agent.{agentId}",
@@ -33,7 +31,6 @@
       "id": "verify",
       "kind": "verify",
       "description": "Verification pipeline: runs stages.json verificationStages in order (tier \"quick\" by default; --full runs the whole list)",
-      "command": "./.har/verify.sh {agentId}",
       "acceptsArgs": [
         "--full"
       ],
@@ -49,7 +46,6 @@
       "id": "status",
       "kind": "inspect",
       "description": "Inspect agent slot status, simulator, and worktree path",
-      "command": "./.har/agent-cli.sh {agentId} status",
       "artifacts": [
         {
           "path": ".har/logs",
@@ -61,8 +57,7 @@
     {
       "id": "teardown",
       "kind": "teardown",
-      "description": "Remove worktree and env file for the slot",
-      "command": "./.har/teardown.sh {agentId}"
+      "description": "Remove worktree and env file for the slot"
     },
     {
       "id": "build",
@@ -85,7 +80,7 @@
       "kind": "test",
       "tier": "full",
       "description": "SwiftLint (skips when not installed)",
-      "command": "command -v \"${HARNESS_SWIFTLINT_CMD:-swiftlint}\" >/dev/null 2>&1 && \"${HARNESS_SWIFTLINT_CMD:-swiftlint}\" --quiet || echo \"swiftlint not installed — skipping\"",
+      "command": "command -v \"${HARNESS_SWIFTLINT_CMD:-swiftlint}\" >/dev/null 2>&1 && \"${HARNESS_SWIFTLINT_CMD:-swiftlint}\" --quiet || echo \"swiftlint not installed \u2014 skipping\"",
       "requiresAgentId": true
     },
     {

package/dist/templates/migration-prompt.md

@@ -9,7 +9,7 @@ In 1.0 the harness machinery lives once, in the `@osfactory/har` package. `.har/
 - **hooks** — `.har/hooks/{pre-launch,post-launch,pre-verify,pre-teardown,post-teardown}.sh` (custom lifecycle behavior)
 - **plugins** — `.har/plugins/<id>/` local plugins for anything bigger (`har plugin create <id>`)
 
-The `./.har/*.sh` entry points become thin shims (`exec har env … "$@"` with a pinned `npx` fallback), so every surface — raw scripts, `har env`, MCP tools — writes the same run/validation records. Your muscle memory (`./.har/launch.sh 1`, `./.har/verify.sh 1 --full`, `har env launch 1`) keeps working before, during, and after.
+Lifecycle wrappers under `.har/` are deleted. Drive the harness with `har env …` or MCP — both write the same run/validation records. Project-owned `stages/*.sh` and `hooks/*.sh` stay.
 
 {{MIGRATION_SECTION}}
 
@@ -35,7 +35,7 @@ If the backups show deep patches to the machinery itself (not just config/stages
 2. `har env maintain` — drift report should show only files you deliberately adapted.
 3. Full verify on **every configured agent slot** that is in use; at minimum:
    - `har env launch 1`
-   - `har env verify 1 --full` (or `./.har/verify.sh 1 --full` — same records either way)
+   - `har env verify 1 --full`
    - `har env complete 1` (or `har env teardown 1`)
 4. Exercise anything you lifted into hooks/stages (e.g. per-slot data stores exist in the work dir, custom checks run in `--full`).
 

package/dist/templates/plugins/custom-stage-skeleton.sh

@@ -26,8 +26,8 @@ escape_step_output() {
   printf '%s' "$1" | node -e "let d='';process.stdin.on('data',c=>d+=c);process.stdin.on('end',()=>{const s=d.trim().split('\n').slice(0,50).join('\n');process.stdout.write(JSON.stringify(s))})" 2>/dev/null || echo '""'
 }
 
-ENV_FILE="${ENV_FILE:?No slot env for agent ${AGENT_ID} — run ./.har/launch.sh ${AGENT_ID} first}"
-WORK_DIR="${WORK_DIR:?No slot work dir for agent ${AGENT_ID} — run ./.har/launch.sh ${AGENT_ID} first}"
+ENV_FILE="${ENV_FILE:?No slot env for agent ${AGENT_ID} — run har env launch ${AGENT_ID} first}"
+WORK_DIR="${WORK_DIR:?No slot work dir for agent ${AGENT_ID} — run har env launch ${AGENT_ID} first}"
 ARTIFACTS_DIR="$HARNESS_DIR/artifacts/__STAGE_ID__"
 mkdir -p "$ARTIFACTS_DIR"
 

package/dist/templates/plugins/gitleaks/.har/stages/secrets-scan.sh

@@ -48,8 +48,8 @@ REPORT_PATH="$ARTIFACT_DIR/report.json"
 SCAN_LOG="$ARTIFACT_DIR/gitleaks.log"
 
 # ── Resolve agent env ─────────────────────────────────────────────────────────
-ENV_FILE="${ENV_FILE:?No slot env for agent ${AGENT_ID} — run ./.har/launch.sh ${AGENT_ID} first}"
-WORK_DIR="${WORK_DIR:?No slot work dir for agent ${AGENT_ID} — run ./.har/launch.sh ${AGENT_ID} first}"
+ENV_FILE="${ENV_FILE:?No slot env for agent ${AGENT_ID} — run har env launch ${AGENT_ID} first}"
+WORK_DIR="${WORK_DIR:?No slot work dir for agent ${AGENT_ID} — run har env launch ${AGENT_ID} first}"
 
 GITLEAKS_VERSION="$(gitleaks version 2>/dev/null | head -1 || echo unknown)"
 log "gitleaks ${GITLEAKS_VERSION} — ${MODE} scan of $WORK_DIR"