agent-toolkit-cli 1.0.9__tar.gz → 1.1.0__tar.gz

{agent_toolkit_cli-1.0.9 → agent_toolkit_cli-1.1.0}/CHANGELOG.md

@@ -72,6 +72,68 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Windows compatibility: `os.getuid()` doesn't exist on Windows — wrapped in try/except
   so `agent-toolkit doctor` no longer crashes on Windows runners
 
+## [1.0.10] — 2026-08-04
+
+### Fixed
+- `loop run` now works from pip/uvx installs — `loop-gh-gate` (the gh CLI
+  security shim) is now included in the wheel at `agent_toolkit/loop/loop-gh-gate`
+- Windows compatibility: `os.getuid()` wrapped in try/except AttributeError
+
+### Added  
+- uv workspace structure: `packages/agent-toolkit-cli/` — ready for future packages
+  (`agent-toolkit-server`, `agent-toolkit-mcp`, etc.)
+- Root `pyproject.toml` with `[tool.uv.workspace]` and shared dev tooling
+
+## [1.1.0] — 2026-08-04
+
+### Added — Full workspace capability
+
+agent-toolkit is now a complete workspace toolkit. Every capability from
+~/.ai-workspace and agentic-harness is available as a CLI subcommand:
+
+**`agent-toolkit workspace`** — Workspace scaffolding and session context
+- `workspace init [--dir PATH]` — scaffold harness workspace (AGENTS.md, knowledge/, packs/, personas/, loops/, projects/, repos/)
+- `workspace context` — session state snapshot for AI session start
+- `workspace sync` — sync loop escalations into knowledge todos
+
+**`agent-toolkit memory`** — Persistent knowledge base (replaces bin/assistant-memory)
+- `memory add --type <learning|process|todo> "content"` — add to knowledge base
+- `memory search "query"` — search all knowledge files
+- `memory inject` — output full knowledge context block for AI session
+- `memory review` — show stale entries
+- `memory todo` — list pending todos
+
+**`agent-toolkit project`** — Project clone and symlink manager (replaces bin/project-indexer)
+- `project clone owner/repo` — clone + symlink into projects/
+- `project list` — list indexed projects
+- `project add <path>` — symlink existing repo
+- `project remove <name>` — remove symlink
+- `project scan` — consistency check
+
+**`agent-toolkit devcompanion`** — Background job queue (replaces bin/devcompanion)
+- `devcompanion queue <project> [--request "..."] [--template NAME]` — queue a job
+- `devcompanion run-once [--no-llm]` — execute oldest pending job
+- `devcompanion status` — show all jobs
+- `devcompanion done <job-id>` — mark job complete
+- `devcompanion sync-todos` — sync plan.md todos to knowledge
+
+**`agent-toolkit insights`** — AI tool usage analytics (replaces bin/tool-insights)
+- `insights opencode` — OpenCode sessions from SQLite DB
+- `insights cursor` — Cursor agent transcripts
+- `insights claude` — Claude Code sessions
+- `insights all` — aggregate across all tools
+
+**Templates** — workspace scaffold templates bundled in the wheel
+- 12 CLI modules, 42 template files (AGENTS.md, personas, knowledge, etc.)
+- `agent-toolkit workspace init` = what agentic-harness now scaffolds from
+
+**Profile updates** — Claude Code, Cursor, OpenCode, Windsurf profiles updated with
+agent-toolkit CLI reference and session-start protocol
+
+### Changed
+- agentic-harness: bin/workspace-context and bin/assistant-memory are now
+  thin wrappers that delegate to agent-toolkit CLI
+
 ## [Unreleased]
 
 ### Added

{agent_toolkit_cli-1.0.9 → agent_toolkit_cli-1.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: agent-toolkit-cli
-Version: 1.0.9
+Version: 1.1.0
 Summary: Composable AI agent toolkit — loops, skills, profiles, and MCP for Claude Code, Cursor, OpenCode, Windsurf, and more
 Project-URL: Homepage, https://github.com/ulises-jeremias/agent-toolkit
 Project-URL: Repository, https://github.com/ulises-jeremias/agent-toolkit.git

agent_toolkit_cli-1.1.0/loops/daily-triage/STATE.md

@@ -0,0 +1,8 @@
+---
+last_run: "2026-08-04T16:40:00Z"
+last_run_status: "completed"
+last_run_id: "20260804T163838Z-548bdc"
+runs_today: 1
+pending: "[]"
+escalations: "[]"
+---

agent_toolkit_cli-1.1.0/loops/daily-triage/runs/20260804T163838Z-548bdc/.gate/bin/gh

@@ -0,0 +1,3 @@
+#!/usr/bin/env bash
+set -euo pipefail
+exec "/tmp/gate-test/bin/python3" "/tmp/gate-test/lib/python3.14/site-packages/agent_toolkit/loop/loop-gh-gate" -- "$@"

agent_toolkit_cli-1.1.0/loops/daily-triage/runs/20260804T163838Z-548bdc/plan.md

@@ -0,0 +1,21 @@
+# Plan — daily-triage run 20260804T163838Z
+
+## Objective
+Review all open issues created in the last 24 hours on `ulises-jeremias/agent-toolkit`,
+propose labels and priority scores, and write a report. No mutations allowed (L1 tier).
+
+## Steps
+
+1. [x] Identify the target repository via `git remote`
+2. [x] Fetch open issues created in the last 24 hours via `gh issue list`
+3. [x] Fetch the existing label taxonomy via `gh label list`
+4. [x] For each issue, assign:
+   - Priority (critical / high / medium / low)
+   - Proposed labels (from existing taxonomy + any already applied)
+   - 1–2 sentence summary
+5. [x] Write `report.md` in the output directory
+6. [x] Update `STATE.md` with pending/escalations
+
+## Constraints
+- Read-only. No `gh` mutations (label, comment, close, merge, push).
+- No security issues found → no escalation required.

agent_toolkit_cli-1.1.0/loops/daily-triage/runs/20260804T163838Z-548bdc/prompt.md

@@ -0,0 +1,34 @@
+You are executing an autonomous loop run in an agent-toolkit workspace.
+Workspace root: /home/ulisesjcf/repos/agent-toolkit
+Loop directory: /home/ulisesjcf/repos/agent-toolkit/loops/daily-triage
+Output directory: /home/ulisesjcf/repos/agent-toolkit/loops/daily-triage/runs/20260804T163838Z-548bdc
+
+## Autonomy contract (HARD — violate = human_escalation)
+- Tier: L1
+- Allowlist (may do without asking): []
+- Deny (never do): merge, close, label, comment, push
+- Verifier for mutating actions: null
+
+Tier rules:
+- L1 report-only: READ and write plan.md/report.md/STATE.md only. Do NOT comment, label, assign, merge, close, push, commit, or approve.
+
+Hard gate (enforced by PATH shim — not honor-system):
+- Mutating `gh` commands are intercepted. Denied actions exit with code 78 (from the intercepted `gh` process).
+- For merge/close: an *independent* verifier must write a receipt JSON (do NOT self-approve as the maker):
+  `$OUTPUT_DIR/verifier-receipts/<slug>.json` with keys:
+  `action`, `repo` (owner/name), `number`, `approved` (true), `verifier` ("null"), `rationale`, `ts` (ISO8601 Z).
+- When `LOOP_GATE_RECEIPT_SECRET` is set, receipts must include `sig` (HMAC-SHA256 hex of the canonical JSON payload).
+- Receipts require exact repo + number + verifier match and expire after 1 hour.
+
+Before finishing:
+1. Update `/home/ulisesjcf/repos/agent-toolkit/loops/daily-triage/STATE.md` frontmatter `pending:` and `escalations:` (replace with current lists; use empty lists if clear). Quote every list item that contains `#` (PR/issue numbers), e.g. `- "nanlabs/foo#123 — reason"`.
+2. Write plan.md and report.md under the output directory.
+3. If exiting with human_escalation, put the reason in escalations and report.md.
+
+---
+
+|
+
+---
+
+Write your final report to /home/ulisesjcf/repos/agent-toolkit/loops/daily-triage/runs/20260804T163838Z-548bdc/report.md and your plan to /home/ulisesjcf/repos/agent-toolkit/loops/daily-triage/runs/20260804T163838Z-548bdc/plan.md. Work from the workspace root shown above.

agent_toolkit_cli-1.1.0/loops/daily-triage/runs/20260804T163838Z-548bdc/report.md

@@ -0,0 +1,81 @@
+# Daily Triage Report
+**Run:** 20260804T163838Z  
+**Repo:** ulises-jeremias/agent-toolkit  
+**Date:** 2026-08-04  
+**Tier:** L1 (report-only — no mutations applied)
+
+---
+
+## Summary
+
+4 open issues were created in the last 24 hours. No security issues found. No escalation needed.
+
+---
+
+## Issue Proposals
+
+### #6 — feat: agent-toolkit loop run — live progress output
+- **Created:** 2026-08-04T15:47:11Z  
+- **Author:** ulises-jeremias  
+- **Current labels:** `enhancement`  
+- **Priority:** medium  
+- **Proposed labels:** `enhancement`, `domain/core`  
+- **Summary:** Proposes streaming real-time output (LLM tokens, progress bar, per-repo status) when running a loop template. A `--quiet` flag would suppress this for scripted use. This is a UX improvement to the core loop orchestration, hence `domain/core`.
+
+---
+
+### #5 — feat: agent-toolkit update — refresh installed profiles
+- **Created:** 2026-08-04T15:47:10Z  
+- **Author:** ulises-jeremias  
+- **Current labels:** `enhancement`  
+- **Priority:** high  
+- **Proposed labels:** `enhancement`, `domain/core`  
+- **Summary:** Adds an `agent-toolkit update` command to refresh installed tool profiles (Claude Code, Cursor, etc.) without a full reinstall, with diff preview and `--check` dry-run support. Directly affects the install/update lifecycle — a core workflow gap.
+
+---
+
+### #4 — feat: shell completion for bash, zsh, fish
+- **Created:** 2026-08-04T15:47:09Z  
+- **Author:** ulises-jeremias  
+- **Current labels:** `enhancement`  
+- **Priority:** low  
+- **Proposed labels:** `enhancement`, `good first issue`  
+- **Summary:** Adds tab completion for subcommands, flags, and tool/loop names across bash, zsh, and fish shells, installable via `agent-toolkit completion <shell>`. Well-scoped and self-contained — a good candidate for a first contributor.
+
+---
+
+### #3 — feat: download data on first run from GitHub Releases (lighter wheel)
+- **Created:** 2026-08-04T15:47:07Z  
+- **Author:** ulises-jeremias  
+- **Current labels:** `enhancement`  
+- **Priority:** medium  
+- **Proposed labels:** `enhancement`, `domain/core`  
+- **Summary:** Proposes lazy-downloading skills/profiles/loops from the latest GitHub Release on first run (XDG cache), reducing the wheel from ~5MB to ~500KB (10x). Offline installs remain possible via `--offline`. Significant infrastructure change touching `_paths.py` and the install/update pipeline.
+
+---
+
+## Escalations
+
+None.
+
+---
+
+## Label Taxonomy Reference
+
+| Label | Description |
+|---|---|
+| `bug` | Something isn't working |
+| `enhancement` | New feature or request |
+| `documentation` | Docs improvements |
+| `good first issue` | Good for newcomers |
+| `help wanted` | Extra attention needed |
+| `loop-request` | New loop template request |
+| `skill-request` | New skill request |
+| `domain/core` | Core orchestration skills |
+| `domain/delivery` | Delivery and work item skills |
+| `domain/design` | Design and Figma skills |
+| `profile/claude-code` | Claude Code profile changes |
+| `profile/cursor` | Cursor profile changes |
+| `profile/opencode` | OpenCode profile changes |
+| `profile/pi` | Pi Coding Agent profile changes |
+| `profile/windsurf` | Windsurf profile changes |

agent_toolkit_cli-1.1.0/loops/daily-triage/runs/20260804T163838Z-548bdc/trace.jsonl

@@ -0,0 +1,3 @@
+{"ts": "2026-08-04T16:38:38Z", "kind": "run_start", "run_id": "20260804T163838Z-548bdc", "tier": "L1", "allowlist": ["[]"], "deny": ["merge", "close", "label", "comment", "push"], "verifier": "null", "agents_md_hash": "73aa9e85edd8", "force": false}
+{"ts": "2026-08-04T16:38:38Z", "kind": "worktree_created", "path": "/home/ulisesjcf/.local/share/agent-toolkit/worktrees/daily-triage/20260804T163838Z-548bdc"}
+{"ts": "2026-08-04T16:40:00Z", "kind": "run_end", "status": "completed"}

{agent_toolkit_cli-1.0.9 → agent_toolkit_cli-1.1.0/packages/agent-toolkit-cli}/src/agent_toolkit/__init__.py

@@ -1,4 +1,4 @@
 """agent-toolkit: Composable AI agent capabilities for all major AI coding tools."""
 
-__version__ = "1.0.9"
+__version__ = "1.1.0"
 __author__ = "ulises-jeremias"