stillwright 0.1.1__py3-none-win_amd64.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
|
Binary file
|
|
@@ -0,0 +1,129 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: stillwright
|
|
3
|
+
Version: 0.1.1
|
|
4
|
+
Classifier: Environment :: Console
|
|
5
|
+
Classifier: Programming Language :: Rust
|
|
6
|
+
Classifier: Topic :: Software Development
|
|
7
|
+
License-File: LICENSE
|
|
8
|
+
Summary: sw, the state keeper for the Stillwright agent skills: goals, runs, inbox, dated notes, hash-pinned tools
|
|
9
|
+
Keywords: agent,orchestrator,ai,skills,cli
|
|
10
|
+
Home-Page: https://github.com/HamzaIlyas/stillwright
|
|
11
|
+
Author: Hamza Ilyas
|
|
12
|
+
License-Expression: MIT
|
|
13
|
+
Requires-Python: >=3.8
|
|
14
|
+
Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
|
|
15
|
+
Project-URL: Changelog, https://github.com/HamzaIlyas/stillwright/blob/main/CHANGELOG.md
|
|
16
|
+
Project-URL: Homepage, https://github.com/HamzaIlyas/stillwright
|
|
17
|
+
|
|
18
|
+
# Stillwright
|
|
19
|
+
|
|
20
|
+
[](https://github.com/HamzaIlyas/stillwright/actions/workflows/ci.yml)
|
|
21
|
+
[](https://github.com/HamzaIlyas/stillwright/releases/latest)
|
|
22
|
+
[](LICENSE)
|
|
23
|
+
|
|
24
|
+
|
|
25
|
+
|
|
26
|
+
**A quiet maker for your AI agent.** It gets to know you, then keeps working on your goals:
|
|
27
|
+
plans them, finds and vets the right tools, proves every step, and only speaks when it
|
|
28
|
+
needs you, in five lines or fewer.
|
|
29
|
+
|
|
30
|
+
Any goal with a way to check it: code, research, writing, launches, operations. It runs
|
|
31
|
+
inside the agent you already use: Claude Code, Codex, Gemini CLI, OpenCode and any agent
|
|
32
|
+
that reads [Agent Skills](https://agentskills.io).
|
|
33
|
+
|
|
34
|
+
## Why
|
|
35
|
+
|
|
36
|
+
- **Agents say "done" when it isn't.** Stillwright passes a step only when an independent
|
|
37
|
+
check proves it: tests run, page loads, source says so.
|
|
38
|
+
- **There are too many tools.** Thousands of skills, MCP servers and repos, some malicious.
|
|
39
|
+
Stillwright picks by evidence, reads the code before trusting it, asks you once, and pins
|
|
40
|
+
it by hash so it can't change silently. If nothing fits, it builds the tool itself.
|
|
41
|
+
- **Agents talk too much.** One inbox (ask, review, notify, idea). Silence is the default.
|
|
42
|
+
|
|
43
|
+
## Install
|
|
44
|
+
|
|
45
|
+
| Agent | Command |
|
|
46
|
+
|---|---|
|
|
47
|
+
| Claude Code | `claude plugin marketplace add HamzaIlyas/stillwright` then `claude plugin install stillwright@stillwright` |
|
|
48
|
+
| Codex | `codex plugin marketplace add HamzaIlyas/stillwright`, then install Stillwright from `/plugins` |
|
|
49
|
+
| Gemini CLI | `gemini extensions install https://github.com/HamzaIlyas/stillwright` |
|
|
50
|
+
| Cursor | loads the root `plugin.json` (Agent Plugins standard); marketplace listing pending |
|
|
51
|
+
| OpenCode, or any Agent Skills agent | `npx skills add HamzaIlyas/stillwright` |
|
|
52
|
+
|
|
53
|
+
The skills need the `sw` binary (a single file, no runtime). Any of these installs it:
|
|
54
|
+
|
|
55
|
+
```sh
|
|
56
|
+
# Linux / macOS
|
|
57
|
+
curl --proto '=https' --tlsv1.2 -LsSf https://github.com/HamzaIlyas/stillwright/releases/latest/download/stillwright-installer.sh | sh
|
|
58
|
+
# Windows (PowerShell)
|
|
59
|
+
powershell -ExecutionPolicy Bypass -c "irm https://github.com/HamzaIlyas/stillwright/releases/latest/download/stillwright-installer.ps1 | iex"
|
|
60
|
+
# From source (needs Rust)
|
|
61
|
+
cargo install --git https://github.com/HamzaIlyas/stillwright --locked
|
|
62
|
+
```
|
|
63
|
+
|
|
64
|
+
On Linux and macOS the skills can also install it on first use (`skills/stillwright/scripts/get-sw.sh`, pinned
|
|
65
|
+
to a release version), when you are present to approve it.
|
|
66
|
+
|
|
67
|
+
## Use
|
|
68
|
+
|
|
69
|
+
New here? The [quickstart](docs/quickstart.md) takes ten minutes, and CI runs every command in it.
|
|
70
|
+
Also: [troubleshooting](docs/troubleshooting.md), [Windows notes](docs/windows.md), and a
|
|
71
|
+
[real example run](examples/quickstart-run/README.md).
|
|
72
|
+
|
|
73
|
+
Tell your agent a goal, or just say **"stillwright"**. The first time, it asks a few short
|
|
74
|
+
questions to get to know you. After that:
|
|
75
|
+
|
|
76
|
+
```
|
|
77
|
+
sw status what's going on, in three lines
|
|
78
|
+
sw inbox list the only things that need you
|
|
79
|
+
sw goal add "..." give it more work
|
|
80
|
+
```
|
|
81
|
+
|
|
82
|
+
To keep it working while you're away, schedule one loop per hour, for example
|
|
83
|
+
`claude -p "Use the stillwright skill: run one loop, then stop."` from cron. Unattended runs
|
|
84
|
+
never push, deploy, publish or install anything new; those wait in your inbox.
|
|
85
|
+
|
|
86
|
+
## What's inside
|
|
87
|
+
|
|
88
|
+
| Skill | Does |
|
|
89
|
+
|---|---|
|
|
90
|
+
| `stillwright` | the loop: pick goal → plan → get tools → execute → verify → report |
|
|
91
|
+
| `sw-onboard` | learns who you are and what you want, in three short rounds at most |
|
|
92
|
+
| `sw-scout` | finds, vets, trials and pins the best existing tool for a job |
|
|
93
|
+
| `sw-forge` | builds and tests a new skill when nothing good exists |
|
|
94
|
+
| `sw-verify` | independent, default-fail proof for every step |
|
|
95
|
+
| `sw-radar` | spots what matters to you (updates, risks, new tools) as rare idea cards |
|
|
96
|
+
| `sw-deep-search` | research that ends in verified, sourced answers |
|
|
97
|
+
| `sw-context-diet` | keeps long sessions lean and cheap |
|
|
98
|
+
| `sw-launch` | pitch, README, posts and launch plan for what you build |
|
|
99
|
+
|
|
100
|
+
All state is plain files in `~/.stillwright` (profile, goals, inbox, notes, pinned tools,
|
|
101
|
+
run logs). Nothing leaves your machine except what your agent already sends.
|
|
102
|
+
|
|
103
|
+
`sw` is a single Rust binary, so there is nothing to install besides it, and it is ready to
|
|
104
|
+
become a background service later.
|
|
105
|
+
|
|
106
|
+
## Status
|
|
107
|
+
|
|
108
|
+
v0.1. Each rule answers a failure that has been measured or reported, not a hunch:
|
|
109
|
+
|
|
110
|
+
- Agents mark work done without testing it: Anthropic's
|
|
111
|
+
[notes on long-running agent harnesses](https://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents).
|
|
112
|
+
Hence independent, default-fail checks.
|
|
113
|
+
- Agents rarely ask for help at the right moment and fill gaps with wrong assumptions:
|
|
114
|
+
[HiL-Bench](https://arxiv.org/abs/2604.09408). Hence the typed inbox with silence as default.
|
|
115
|
+
- Agent success falls as tasks get longer; METR measures how long a task an agent completes
|
|
116
|
+
at 50% and at 80% reliability, and the 80% horizon is much shorter:
|
|
117
|
+
[METR time horizons](https://metr.org/time-horizons/). Hence small, checkable steps.
|
|
118
|
+
- Hundreds of malicious skills were found on a public skill registry in early 2026 (341 of 2,857 scanned):
|
|
119
|
+
[Snyk, "From SKILL.md to Shell Access"](https://snyk.io/articles/skill-md-shell-access/),
|
|
120
|
+
[Unit 42 on the OpenClaw skill marketplace](https://unit42.paloaltonetworks.com/openclaw-ai-supply-chain-risk/).
|
|
121
|
+
Hence vetting, approval and hash-pinning before any third-party tool runs.
|
|
122
|
+
- In a controlled test with a limited verification budget, agents followed a constraint that
|
|
123
|
+
had since been withdrawn in about three of four decisions:
|
|
124
|
+
[arXiv 2608.25553](https://arxiv.org/abs/2608.25553). Hence dated notes that are retired, not overwritten.
|
|
125
|
+
|
|
126
|
+
Issues and PRs welcome. Found a gap? Open an [idea](https://github.com/HamzaIlyas/stillwright/issues/new?template=idea.yml).
|
|
127
|
+
|
|
128
|
+
MIT © Hamza Ilyas · Built at [Folvix](https://huggingface.co/Folvix).
|
|
129
|
+
|
|
@@ -0,0 +1,6 @@
|
|
|
1
|
+
stillwright-0.1.1.data/scripts/sw.exe,sha256=eb13kqn57zANJmgeUZ-uacmBIF3nwYNSM5fZ_V6Zt18,1161216
|
|
2
|
+
stillwright-0.1.1.dist-info/METADATA,sha256=Nl4QW7WUwbtCcHKJCzVaSPPmz5cxzwX8chmXfJnAacc,6898
|
|
3
|
+
stillwright-0.1.1.dist-info/WHEEL,sha256=8Aej0W0a6Cz6apA3IzJrTnxLRVLAt-w0Oh8SA3Con_c,94
|
|
4
|
+
stillwright-0.1.1.dist-info/licenses/LICENSE,sha256=0s3Xgog3frB__G6V4syI99XWXAXr-kR2KB-9Cd3cUg8,1089
|
|
5
|
+
stillwright-0.1.1.dist-info/sboms/stillwright.cyclonedx.json,sha256=EZbxuDqxGi51Q0l_GrwvDHWZtE8aQl0j8Jytta5bXME,44890
|
|
6
|
+
stillwright-0.1.1.dist-info/RECORD,,
|
|
@@ -0,0 +1,21 @@
|
|
|
1
|
+
MIT License
|
|
2
|
+
|
|
3
|
+
Copyright (c) 2026 Hamza Ilyas
|
|
4
|
+
|
|
5
|
+
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
6
|
+
of this software and associated documentation files (the "Software"), to deal
|
|
7
|
+
in the Software without restriction, including without limitation the rights
|
|
8
|
+
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
9
|
+
copies of the Software, and to permit persons to whom the Software is
|
|
10
|
+
furnished to do so, subject to the following conditions:
|
|
11
|
+
|
|
12
|
+
The above copyright notice and this permission notice shall be included in all
|
|
13
|
+
copies or substantial portions of the Software.
|
|
14
|
+
|
|
15
|
+
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
16
|
+
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
17
|
+
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
18
|
+
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
19
|
+
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
20
|
+
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
|
|
21
|
+
SOFTWARE.
|