agent-devkit 0.7.0 → 0.8.3

package/README.md

@@ -1,16 +1,16 @@
 # Agent DevKit
 
-Agent DevKit is a TypeScript/Node rewrite of the `agent-devkit` package.
+Agent DevKit is a TypeScript/Node autonomous agent for the terminal.
 
 The public npm package is `agent-devkit`; the canonical command is `agent`.
-Version `0.3.5` starts a new implementation focused on a maintainable CLI/TUI
-runtime while preserving the product direction from the legacy `0.3.x` line.
+The current version is defined in `package.json` — it is the single source
+of truth; this document does not promise a specific release.
 
 ## Status
 
-This branch is the technical foundation for the rewrite. The package setup,
-build, checks and release pipeline are prepared before product behavior is
-implemented.
+Active development. The runtime ships a personality-driven chat and task
+agent (CLI, TUI and MCP), multi-provider brains with fallback, persisted
+runs/goals/memory and multi-agent delegation.
 
 ## Install
 
@@ -145,7 +145,7 @@ Rules:
 ## Distribution
 
 The package is npm-first and publishes the `agent` binary from `dist/main.js`.
-Standalone binaries are outside the initial `0.3.5` scope.
+Standalone binaries are out of scope for now.
 
 Installing the npm package must not create runtime folders or project files.
 State is created only when the user runs the tool:
@@ -158,7 +158,6 @@ agent
   may create ~/.agent-devkit when global state is needed
 
 agent init
-agent install project
   may create <project>/.agent-devkit for project-local state
 ```
 
@@ -187,8 +186,8 @@ agent reset --dry-run
 agent reset --yes
 agent reset -g --dry-run
 agent update --latest --dry-run
-agent update 0.3.5 --dry-run
-agent update 0.3.5 --yes
+agent update <version> --dry-run
+agent update <version> --yes
 agent preferences
 agent preferences --json
 agent preferences themes
@@ -213,8 +212,11 @@ agent tools
 agent tools --json
 agent install node --dry-run
 agent install node --verify
+agent tooling list
+agent tooling plan-install gemini
+agent tooling upgrade claude --yes
 agent run project.doctor --input '{}' --json
-agent run environment.dependencies --input '{"action":"verify","dependency":"node"}' --json
+agent run environment.tooling --input '{"action":"detect","toolingId":"node"}' --json
 agent run project.reset --input '{"confirmed":true,"dryRun":false,"homeDirectory":"/tmp/home","projectRoot":"/tmp/project","scope":"project"}' --approve --json
 agent mcp
 agent mcp stdio
@@ -264,11 +266,22 @@ Risky capabilities return `approval_required` unless `--approve` is passed. MCP
 uses the same runtime contract, so future TUI and agent-loop execution should
 also plug into this layer instead of calling capabilities directly.
 
+### Pack trust model
+
+Packs execute with the full privileges of the `agent` process: `node` runners
+are imported in-process and `command` runners spawn as child processes (no
+shell). Install only packs you trust — the approval gate (required by default)
+is the execution safety net, not a sandbox. `command` runners receive a minimal
+environment (`PATH`, `HOME`, `LANG`) and the JSON input via stdin, so parent
+secrets and provider keys are never inherited.
+
 ## Agent Runtime, Memory and Integrations
 
 The agentic runtime persists every run under `~/.agent-devkit/data/runtime/`
 (`run.json`, `events.jsonl`, `plan.json`) with goals in `data/runtime/goals/`.
 Runs are inspectable and resumable through `agent runs` and `agent goals`.
+Direct capability calls from the TUI and CLI act with the user's own
+privileges; the approval gates protect the agentic path.
 
 Global memory lives in the `memory` module: records have kind
 (`fact`, `preference`, `pattern`, `decision`, `episode`), status
@@ -343,25 +356,27 @@ that is removed before capability validation:
 CLI and MCP logging must not persist raw tool inputs. Sensitive CLI flags such
 as `agent run --input` are redacted before usage and technical logs are written.
 
-## Environment Dependencies
+## Environment Tooling
 
-Agent DevKit has a provider-based dependency lifecycle module exposed as the
-`environment.dependencies` capability.
+Agent DevKit has a catalog-driven tooling lifecycle module exposed as the
+`environment.tooling` capability. CLI, MCP and the agent runtime use this same
+capability to detect, doctor, inspect versions, plan, install, upgrade,
+downgrade and uninstall supported tools.
 
 ```bash
 agent install node --dry-run
 agent install node --verify --json
 agent install --node --verify
+agent tooling list
+agent tooling detect ollama
+agent tooling plan-install gemini
+agent tooling install gemini --yes
+agent run environment.tooling --input '{"action":"detect","toolingId":"node"}' --json
 ```
 
-The first provider is `node`. It is read-only: it checks the current Node.js
-runtime, compatibility and environment, but does not install, upgrade,
-downgrade or uninstall Node.js. Mutating lifecycle actions currently return
-planned or unsupported results without side effects.
-
-Future providers such as `aws-cli`, `docker`, `gh-cli` and `ollama` should
-implement the same provider contract instead of adding one-off installer logic
-to CLI, TUI, MCP or agent loops.
+Official tooling definitions live in `src/assets/tooling/`. User-learned or
+promoted definitions live under `~/.agent-devkit/data/tooling/` and are loaded
+through the same catalog.
 
 ## Internal Docs