Fix/drop target all

.claude/skills/base/skill.md

@@ -5,7 +5,7 @@ triggers:
   alwaysActivate: true
 ---
 
-You are working in **aspens** — a CLI that keeps coding-agent context accurate as your codebase changes. Scans repos, generates project-specific instructions and skills for Claude Code and Codex CLI, and keeps them fresh.
+You are working in **aspens** — a CLI that keeps coding-agent context accurate as your codebase changes. Scans repos, generates project-specific instructions and skills for Claude Code, Codex, and OpenCode CLI, and keeps them fresh.
 
 ## Tech Stack
 Node.js 20+ (ESM) | Commander | Vitest | es-module-lexer | @clack/prompts | picocolors
@@ -14,7 +14,7 @@ Node.js 20+ (ESM) | Commander | Vitest | es-module-lexer | @clack/prompts | pico
 - `npm test` — Run vitest suite
 - `npm start` / `node bin/cli.js` — Run CLI
 - `aspens scan [path]` — Deterministic repo analysis (no LLM)
-- `aspens doc init [path]` — Generate skills + hooks + CLAUDE.md (`--target claude|codex|all`, `--recommended` for full recommended setup)
+- `aspens doc init [path]` — Generate skills + hooks + CLAUDE.md (`--target claude|codex|opencode`, `--recommended` for full recommended setup)
 - `aspens doc impact [path]` — Show freshness, coverage, and drift of generated context (`--apply` for auto-repair, `--backend`/`--model`/`--timeout`/`--verbose` for LLM interpretation)
 - `aspens doc sync [path]` — Incremental skill updates from git diffs
 - `aspens doc graph [path]` — Rebuild import graph cache (`.claude/graph.json`)

CHANGELOG.md

@@ -1,5 +1,21 @@
 # Changelog
 
+## [0.9.0] - 2026-08-15
+
+### Added
+- **OpenCode CLI support** — `opencode` is now a supported generation backend and output target across `doc init` / `doc sync` / `doc impact`. `runOpenCode()` drives `opencode run --format json` (prompt passed via a `-f` file, stdin disabled to avoid a startup hang) and adapts its event stream to the shared `{ text, usage }` runner interface. The target writes `AGENTS.md` + `.claude/skills`. (#48)
+- **`.php` recognized as a code-bearing extension** — PHP source files are now counted by the scanner and domain discovery. (#47)
+
+### Changed
+- **`--target all` removed** — with three targets (and `codex`/`opencode` both owning `AGENTS.md`), `all` was ambiguous. Pick an explicit target: `--target claude|codex|opencode`. Passing `--target all` now errors with the valid list; run `doc init --target <one>` per target instead.
+
+### Fixed
+- **OpenCode runner hardening** — dropped `--dangerously-skip-permissions`; added a safe-charset guard on `--model` before it reaches argv; replaced shell-based temp-file cleanup with `rmSync`; buffered JSON events across stdout chunks so generated text isn't lost; classified rate-limit exits. (#48)
+- **`doc init` target/backend selection generalized** — the backend picker, target multiselect, and `inferConfig()` recovery now handle all three targets instead of hardcoding claude/codex; conflicting shared-output targets (`codex` + `opencode` both writing `AGENTS.md`) are rejected with a clear error. (#48)
+
+### Security
+- **Dev-dependency advisories resolved** — `npm audit fix` patched transitive `postcss`, `vite`, and `nanoid` advisories (dev-only; runtime dependencies were already clean). `npm audit` now reports 0 vulnerabilities.
+
 ## [0.8.0] - 2026-05-11
 
 ### Added

CLAUDE.md

@@ -21,7 +21,7 @@
 - `npm start` — run the CLI (`node bin/cli.js`)
 - `npm run lint` — no-op check (`echo 'No linter configured yet' && exit 0`)
 - `aspens scan [path]` — deterministic repo scan
-- `aspens doc init [path]` — generate skills, hooks, and instructions file (`--target claude|codex|all`, `--recommended` for full recommended setup including save-tokens, agents, and doc-sync hook)
+- `aspens doc init [path]` — generate skills, hooks, and instructions file (`--target claude|codex|opencode`, `--recommended` for full recommended setup including save-tokens, agents, and doc-sync hook)
 - `aspens doc impact [path]` — show freshness, coverage, drift, and LLM interpretation of generated context (interactive apply for repairs)
 - `aspens doc sync [path]` — update docs from recent diffs
 - `aspens doc graph [path]` — rebuild `.claude/graph.json`

README.md

@@ -17,7 +17,7 @@ You started with 50 clean lines. Three months later it's 200, and Claude ignores
 
 aspens replaces the monolith with scoped skill files (~35 lines each) generated from your actual import graph. Each skill activates only when the agent touches that part of the codebase. A post-commit hook keeps them in sync automatically. The agent reads 35 focused lines instead of 200 sprawling ones, and actually follows them.
 
-Works with Claude Code, Codex, or both.
+Works with Claude Code, Codex, or OpenCode.
 
 ## Install
 
@@ -98,7 +98,7 @@ Aspens supports multiple agent environments through output targets:
 
 - `claude`: `CLAUDE.md` + `.claude/skills` + Claude hooks
 - `codex`: `AGENTS.md` + `.agents/skills` + directory `AGENTS.md`
-- `all`: generate both sets together
+- `opencode`: `AGENTS.md` + `.claude/skills`
 - we are working on adding more agents and tools - ask or contribute!
 
 ## Commands
@@ -140,8 +140,8 @@ $ aspens doc init
 | `--no-graph` | Skip import graph analysis |
 | `--model <model>` | Model for the selected backend |
 | `--verbose` | Show backend reads/activity in real time |
-| `--target <target>` | Output target: `claude`, `codex`, or `all` |
-| `--backend <backend>` | Generation backend: `claude` or `codex` |
+| `--target <target>` | Output target: `claude`, `codex`, or `opencode` |
+| `--backend <backend>` | Generation backend: `claude`, `codex`, or `opencode` |
 
 ### `aspens doc impact`
 
@@ -258,6 +258,7 @@ aspens save-tokens --remove      # Uninstall
 - At least one supported agent CLI:
   - [Claude Code CLI](https://docs.anthropic.com/en/docs/claude-code) — `npm install -g @anthropic-ai/claude-code`
   - Codex CLI
+  - [OpenCode CLI](https://opencode.ai)
 
 ## License
 

bin/cli.js

@@ -53,7 +53,6 @@ function showWelcome() {
     ${pc.green('aspens doc init --recommended')}     Install the full recommended setup
     ${pc.green('aspens doc init --dry-run')}         Preview without writing
     ${pc.green('aspens doc init --mode chunked')}    One domain at a time (large repos)
-    ${pc.green('aspens doc init --target all')}      Generate docs for every configured target
     ${pc.green('aspens doc init --model haiku')}     Use a specific backend model
     ${pc.green('aspens doc init --verbose')}         See backend activity in real time
     ${pc.green('aspens doc sync')}                   Update generated docs from recent commits
@@ -79,7 +78,7 @@ function showWelcome() {
     ${pc.yellow('--force')}            Overwrite existing files     ${pc.yellow('--model')} ${pc.dim('<m>')}   Choose backend model
     ${pc.yellow('--mode')} ${pc.dim('<mode>')}       all, chunked, base-only     ${pc.yellow('--timeout')} ${pc.dim('<s>')}  Seconds per call
     ${pc.yellow('--strategy')} ${pc.dim('<s>')}    improve, rewrite, skip      ${pc.yellow('--json')}      JSON output (scan)
-    ${pc.yellow('--target')} ${pc.dim('<t>')}       claude, codex, opencode, all ${pc.yellow('--backend')} ${pc.dim('<b>')} claude, codex, or opencode
+    ${pc.yellow('--target')} ${pc.dim('<t>')}       claude, codex, opencode      ${pc.yellow('--backend')} ${pc.dim('<b>')} claude, codex, or opencode
     ${pc.yellow('--no-hooks')}         Skip Claude hook installation ${pc.yellow('--hooks-only')}  Update Claude hooks only
     ${pc.yellow('--no-graph')}         Skip import graph analysis
 
@@ -165,7 +164,7 @@ doc
   .option('--no-hooks', 'Skip Claude hook/rules/settings installation')
   .option('--hooks-only', 'Skip doc generation, just install/update Claude hooks')
   .option('--no-graph', 'Skip import graph analysis')
-  .option('--target <target>', 'Output target: claude, codex, opencode, all')
+  .option('--target <target>', 'Output target: claude, codex, opencode')
   .option('--backend <backend>', 'Generation backend: claude, codex, opencode (default: matches target)')
   .action(docInitCommand);