@saifymatteo/pi-extensions 0.1.0 → 0.1.2
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.
- package/README.md +38 -32
- package/package.json +6 -2
- package/packages/pi-bg-bash/README.md +4 -0
- package/packages/pi-bg-bash/package.json +1 -1
- package/packages/pi-claude-usage/README.md +5 -1
- package/packages/pi-claude-usage/package.json +1 -1
- package/packages/pi-claude-usage/src/index.ts +10 -1
- package/packages/pi-claude-usage/test/behavior.test.mjs +71 -0
- package/packages/pi-clipboard-image/README.md +4 -0
- package/packages/pi-clipboard-image/package.json +1 -1
- package/packages/pi-codeblock-borders/README.md +4 -0
- package/packages/pi-codeblock-borders/package.json +1 -1
- package/packages/pi-codegraph/README.md +4 -0
- package/packages/pi-codegraph/package.json +1 -1
- package/packages/pi-compact-tools/README.md +4 -0
- package/packages/pi-compact-tools/package.json +1 -1
- package/packages/pi-decisions/README.md +4 -0
- package/packages/pi-decisions/package.json +1 -1
- package/packages/pi-load-pi-dotenv/README.md +4 -0
- package/packages/pi-load-pi-dotenv/package.json +1 -1
- package/packages/pi-model-go/README.md +4 -0
- package/packages/pi-model-go/package.json +1 -1
- package/packages/pi-prestart-working/README.md +4 -0
- package/packages/pi-prestart-working/package.json +1 -1
- package/packages/pi-root-guard/README.md +4 -0
- package/packages/pi-root-guard/package.json +1 -1
- package/packages/pi-session-finder/README.md +4 -0
- package/packages/pi-session-finder/package.json +1 -1
- package/packages/pi-session-id/README.md +4 -0
- package/packages/pi-session-id/package.json +1 -1
- package/packages/pi-session-nav/README.md +4 -0
- package/packages/pi-session-nav/package.json +1 -1
- package/packages/pi-session-rename/README.md +4 -0
- package/packages/pi-session-rename/package.json +1 -1
- package/packages/pi-session-retention/README.md +4 -0
- package/packages/pi-session-retention/package.json +1 -1
- package/packages/pi-skill-usage-log/README.md +4 -0
- package/packages/pi-skill-usage-log/package.json +1 -1
- package/packages/pi-tool-usage-log/README.md +4 -0
- package/packages/pi-tool-usage-log/package.json +1 -1
- package/packages/pi-tts/README.md +4 -0
- package/packages/pi-tts/package.json +1 -1
- package/packages/pi-user-image-previews/README.md +4 -0
- package/packages/pi-user-image-previews/package.json +1 -1
- package/packages/pi-vision/README.md +5 -1
- package/packages/pi-vision/package.json +1 -1
- package/packages/pi-vision/skills/vision-tools/SKILL.md +37 -21
- package/packages/pi-vision/skills/vision-tools/references/long-screenshot-ocr.md +7 -7
- package/packages/pi-vision/skills/vision-tools/references/restore-graphic.md +7 -7
- package/packages/pi-vision/skills/vision-tools/references/restore-ui.md +4 -4
- package/packages/pi-vision/skills/vision-tools/scripts/long_screenshot_ocr.py +4 -4
- package/packages/pi-vision/test/skill-cli.test.mjs +71 -0
- package/packages/pi-wait/README.md +4 -0
- package/packages/pi-wait/package.json +1 -1
- package/packages/pi-zvec/README.md +4 -0
- package/packages/pi-zvec/package.json +1 -1
package/README.md
CHANGED
|
@@ -2,8 +2,12 @@
|
|
|
2
2
|
|
|
3
3
|
Reusable pieces of a personal Pi setup, packaged separately from the setup itself.
|
|
4
4
|
|
|
5
|
+
|
|
6
|
+
|
|
5
7
|
**Work in progress.** Packages are licensed AGPL-3.0-only ([NOTICE.md](NOTICE.md)) and publish through GitHub Actions when tagged: each extension individually, plus the root as the all-in-one collection [`@saifymatteo/pi-extensions`](https://github.com/saifymatteo/pi-extensions). Git is local for now, with no GitHub repository yet.
|
|
6
8
|
|
|
9
|
+
Each package README opens with a screenshot of what it adds. The screenshots are real captures, not mockups: stock Pi 1.4.0 driven by a scripted demo model, in an isolated agent home. The catalogue below shows them as thumbnails; click one for the package README.
|
|
10
|
+
|
|
7
11
|
## Start here
|
|
8
12
|
|
|
9
13
|
Use Node.js 24 or newer. From this repository:
|
|
@@ -27,6 +31,8 @@ Host-provided Pi packages (`@earendil-works/*`) are declared only as package pee
|
|
|
27
31
|
- `pack:check`: asks npm to show each standalone package's and the root collection's prospective contents without publishing or creating archives.
|
|
28
32
|
- `smoke`: loads each package through isolated host help/startup. Checks module/factory loading, not session events, interactive rendering, models, external CLIs, or backend compatibility. The host is located automatically (`PI_HOST`, `pi` on PATH, then `pi-bolt` for a Pi-Bolt build, then the local `~/.pi-bolt` install). Pi-Bolt is optional; set `PI_HOST` to another executable to compare hosts. A pass here does not prove compatibility with a different host build.
|
|
29
33
|
- `context`: measures what each package adds to the first-turn context window. Current result: the whole collection costs about **1.7k tokens** on top of stock Pi, and 17 of 22 packages contribute nothing (they register hooks, commands, or UI patches only). Measured by capturing the actual request Pi sends to a local provider, against the pinned stock host, with machine state excluded. First-turn static cost only: runtime injections (tool results, hook-driven messages) are not measured. Not part of the `check` gate.
|
|
34
|
+
- `screenshots`: regenerates each `packages/<name>/screenshot.png` (`npm run screenshots -- pi-bg-bash` for one). Drives the pinned stock host in a pseudo-terminal against a scripted local model and fake service endpoints, in a throwaway `/tmp/demo` home, with a stubbed clipboard. macOS only for now (font path, clipboard stub). Needs `agent-browser` on PATH. One-time setup: `npm ci --prefix scripts/screenshots`, then make node-pty's helper executable (`chmod +x scripts/screenshots/node_modules/node-pty/prebuilds/*/spawn-helper`) if npm skipped its install script. Not part of the `check` gate.
|
|
35
|
+
- `demo`: re-records the root `demo.gif` (also the gallery preview via `pi.image`): one session with the whole root collection loaded, same harness and setup as `screenshots`. `DRY=1` prints each scene's screen without rendering. Not part of the `check` gate.
|
|
30
36
|
|
|
31
37
|
## Isolated sessions
|
|
32
38
|
|
|
@@ -50,46 +56,46 @@ Each directory under `packages/` is an independently loadable, independently pub
|
|
|
50
56
|
|
|
51
57
|
### Session and model navigation
|
|
52
58
|
|
|
53
|
-
| Package | Purpose |
|
|
54
|
-
| --- | --- |
|
|
55
|
-
| [pi-session-id](packages/pi-session-id) | Copy the current session ID |
|
|
56
|
-
| [pi-session-finder](packages/pi-session-finder) | Search saved sessions across projects; open or fork them |
|
|
57
|
-
| [pi-session-nav](packages/pi-session-nav) | Jump between messages in the active context (compaction-aware) |
|
|
58
|
-
| [pi-session-rename](packages/pi-session-rename) | Generate session titles from the first prompt using configured models |
|
|
59
|
-
| [pi-session-retention](packages/pi-session-retention) | Inspect and manage stored sessions; review deletion behavior before use |
|
|
60
|
-
| [pi-model-go](packages/pi-model-go) | `/model-go`: cost-aware model picker with Go-plan pricing and intelligence scores |
|
|
59
|
+
| Package | Purpose | Preview |
|
|
60
|
+
| --- | --- | --- |
|
|
61
|
+
| [pi-session-id](packages/pi-session-id) | Copy the current session ID | <a href="packages/pi-session-id"><img src="packages/pi-session-id/screenshot.png" width="280" alt="pi-session-id screenshot"></a> |
|
|
62
|
+
| [pi-session-finder](packages/pi-session-finder) | Search saved sessions across projects; open or fork them | <a href="packages/pi-session-finder"><img src="packages/pi-session-finder/screenshot.png" width="280" alt="pi-session-finder screenshot"></a> |
|
|
63
|
+
| [pi-session-nav](packages/pi-session-nav) | Jump between messages in the active context (compaction-aware) | <a href="packages/pi-session-nav"><img src="packages/pi-session-nav/screenshot.png" width="280" alt="pi-session-nav screenshot"></a> |
|
|
64
|
+
| [pi-session-rename](packages/pi-session-rename) | Generate session titles from the first prompt using configured models | <a href="packages/pi-session-rename"><img src="packages/pi-session-rename/screenshot.png" width="280" alt="pi-session-rename screenshot"></a> |
|
|
65
|
+
| [pi-session-retention](packages/pi-session-retention) | Inspect and manage stored sessions; review deletion behavior before use | <a href="packages/pi-session-retention"><img src="packages/pi-session-retention/screenshot.png" width="280" alt="pi-session-retention screenshot"></a> |
|
|
66
|
+
| [pi-model-go](packages/pi-model-go) | `/model-go`: cost-aware model picker with Go-plan pricing and intelligence scores | <a href="packages/pi-model-go"><img src="packages/pi-model-go/screenshot.png" width="280" alt="pi-model-go screenshot"></a> |
|
|
61
67
|
|
|
62
68
|
### Transcript and terminal UI
|
|
63
69
|
|
|
64
|
-
| Package | Purpose |
|
|
65
|
-
| --- | --- |
|
|
66
|
-
| [pi-compact-tools](packages/pi-compact-tools) | Group consecutive tool calls into one collapsible card |
|
|
67
|
-
| [pi-clipboard-image](packages/pi-clipboard-image) | Attach clipboard images or copied image files (Alt+V) |
|
|
68
|
-
| [pi-codeblock-borders](packages/pi-codeblock-borders) | Render code fences as rules with titles, line numbers, and highlights |
|
|
69
|
-
| [pi-user-image-previews](packages/pi-user-image-previews) | Collapsible inline previews for newly submitted attachments |
|
|
70
|
-
| [pi-claude-usage](packages/pi-claude-usage) | Show Claude 5h/7d subscription usage in the footer |
|
|
71
|
-
| [pi-prestart-working](packages/pi-prestart-working) | Labelled "Working" spinner while `before_agent_start` hooks run |
|
|
70
|
+
| Package | Purpose | Preview |
|
|
71
|
+
| --- | --- | --- |
|
|
72
|
+
| [pi-compact-tools](packages/pi-compact-tools) | Group consecutive tool calls into one collapsible card | <a href="packages/pi-compact-tools"><img src="packages/pi-compact-tools/screenshot.png" width="280" alt="pi-compact-tools screenshot"></a> |
|
|
73
|
+
| [pi-clipboard-image](packages/pi-clipboard-image) | Attach clipboard images or copied image files (Alt+V) | <a href="packages/pi-clipboard-image"><img src="packages/pi-clipboard-image/screenshot.png" width="280" alt="pi-clipboard-image screenshot"></a> |
|
|
74
|
+
| [pi-codeblock-borders](packages/pi-codeblock-borders) | Render code fences as rules with titles, line numbers, and highlights | <a href="packages/pi-codeblock-borders"><img src="packages/pi-codeblock-borders/screenshot.png" width="280" alt="pi-codeblock-borders screenshot"></a> |
|
|
75
|
+
| [pi-user-image-previews](packages/pi-user-image-previews) | Collapsible inline previews for newly submitted attachments | <a href="packages/pi-user-image-previews"><img src="packages/pi-user-image-previews/screenshot.png" width="280" alt="pi-user-image-previews screenshot"></a> |
|
|
76
|
+
| [pi-claude-usage](packages/pi-claude-usage) | Show Claude 5h/7d subscription usage in the footer | <a href="packages/pi-claude-usage"><img src="packages/pi-claude-usage/screenshot.png" width="280" alt="pi-claude-usage screenshot"></a> |
|
|
77
|
+
| [pi-prestart-working](packages/pi-prestart-working) | Labelled "Working" spinner while `before_agent_start` hooks run | <a href="packages/pi-prestart-working"><img src="packages/pi-prestart-working/screenshot.png" width="280" alt="pi-prestart-working screenshot"></a> |
|
|
72
78
|
|
|
73
79
|
### Tools and service adapters
|
|
74
80
|
|
|
75
|
-
| Package | Purpose |
|
|
76
|
-
| --- | --- |
|
|
77
|
-
| [pi-bg-bash](packages/pi-bg-bash) | Add background execution to the built-in bash/powershell tools |
|
|
78
|
-
| [pi-wait](packages/pi-wait) | Join background jobs (shell:/agent:) with lease-based delivery |
|
|
79
|
-
| [pi-codegraph](packages/pi-codegraph) | CodeGraph CLI adapter: explore tool plus index auto-init/sync |
|
|
80
|
-
| [pi-zvec](packages/pi-zvec) | Workspace search via the zg CLI: exact, lexical, semantic, hybrid |
|
|
81
|
-
| [pi-decisions](packages/pi-decisions) | Agent-facing classify tool backed by configured Decisions endpoints |
|
|
82
|
-
| [pi-tts](packages/pi-tts) | Generate WAV files through an OpenAI-compatible speech endpoint |
|
|
83
|
-
| [pi-vision](packages/pi-vision) | Describe images for text-only models via a configured vision chain |
|
|
81
|
+
| Package | Purpose | Preview |
|
|
82
|
+
| --- | --- | --- |
|
|
83
|
+
| [pi-bg-bash](packages/pi-bg-bash) | Add background execution to the built-in bash/powershell tools | <a href="packages/pi-bg-bash"><img src="packages/pi-bg-bash/screenshot.png" width="280" alt="pi-bg-bash screenshot"></a> |
|
|
84
|
+
| [pi-wait](packages/pi-wait) | Join background jobs (shell:/agent:) with lease-based delivery | <a href="packages/pi-wait"><img src="packages/pi-wait/screenshot.png" width="280" alt="pi-wait screenshot"></a> |
|
|
85
|
+
| [pi-codegraph](packages/pi-codegraph) | CodeGraph CLI adapter: explore tool plus index auto-init/sync | <a href="packages/pi-codegraph"><img src="packages/pi-codegraph/screenshot.png" width="280" alt="pi-codegraph screenshot"></a> |
|
|
86
|
+
| [pi-zvec](packages/pi-zvec) | Workspace search via the zg CLI: exact, lexical, semantic, hybrid | <a href="packages/pi-zvec"><img src="packages/pi-zvec/screenshot.png" width="280" alt="pi-zvec screenshot"></a> |
|
|
87
|
+
| [pi-decisions](packages/pi-decisions) | Agent-facing classify tool backed by configured Decisions endpoints | <a href="packages/pi-decisions"><img src="packages/pi-decisions/screenshot.png" width="280" alt="pi-decisions screenshot"></a> |
|
|
88
|
+
| [pi-tts](packages/pi-tts) | Generate WAV files through an OpenAI-compatible speech endpoint | <a href="packages/pi-tts"><img src="packages/pi-tts/screenshot.png" width="280" alt="pi-tts screenshot"></a> |
|
|
89
|
+
| [pi-vision](packages/pi-vision) | Describe images for text-only models via a configured vision chain | <a href="packages/pi-vision"><img src="packages/pi-vision/screenshot.png" width="280" alt="pi-vision screenshot"></a> |
|
|
84
90
|
|
|
85
91
|
### Environment and local observability
|
|
86
92
|
|
|
87
|
-
| Package | Purpose |
|
|
88
|
-
| --- | --- |
|
|
89
|
-
| [pi-load-pi-dotenv](packages/pi-load-pi-dotenv) | Load the agent-local .env into the environment for tool subprocesses |
|
|
90
|
-
| [pi-root-guard](packages/pi-root-guard) | Block agent writes into the Pi and Claude config roots |
|
|
91
|
-
| [pi-skill-usage-log](packages/pi-skill-usage-log) | Log which skills were actually used; `/skill-usage` summary |
|
|
92
|
-
| [pi-tool-usage-log](packages/pi-tool-usage-log) | Passive log of tool executions (names and status, never contents) |
|
|
93
|
+
| Package | Purpose | Preview |
|
|
94
|
+
| --- | --- | --- |
|
|
95
|
+
| [pi-load-pi-dotenv](packages/pi-load-pi-dotenv) | Load the agent-local .env into the environment for tool subprocesses | <a href="packages/pi-load-pi-dotenv"><img src="packages/pi-load-pi-dotenv/screenshot.png" width="280" alt="pi-load-pi-dotenv screenshot"></a> |
|
|
96
|
+
| [pi-root-guard](packages/pi-root-guard) | Block agent writes into the Pi and Claude config roots | <a href="packages/pi-root-guard"><img src="packages/pi-root-guard/screenshot.png" width="280" alt="pi-root-guard screenshot"></a> |
|
|
97
|
+
| [pi-skill-usage-log](packages/pi-skill-usage-log) | Log which skills were actually used; `/skill-usage` summary | <a href="packages/pi-skill-usage-log"><img src="packages/pi-skill-usage-log/screenshot.png" width="280" alt="pi-skill-usage-log screenshot"></a> |
|
|
98
|
+
| [pi-tool-usage-log](packages/pi-tool-usage-log) | Passive log of tool executions (names and status, never contents) | <a href="packages/pi-tool-usage-log"><img src="packages/pi-tool-usage-log/screenshot.png" width="280" alt="pi-tool-usage-log screenshot"></a> |
|
|
93
99
|
|
|
94
100
|
## Individual and collection loading
|
|
95
101
|
|
|
@@ -133,7 +139,7 @@ packages/<extension>/
|
|
|
133
139
|
src/index.ts # existing extension implementation
|
|
134
140
|
src/lib/ # only helpers needed by this package, where applicable
|
|
135
141
|
examples/ # sanitized, inert examples where applicable
|
|
136
|
-
scripts/ # boundary checks, isolated sessions, host smoke
|
|
142
|
+
scripts/ # boundary checks, isolated sessions, host smoke, screenshots
|
|
137
143
|
test/ # test-only TS loader and collection behavior checks
|
|
138
144
|
inventory.json # inclusion/exclusion map
|
|
139
145
|
```
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@saifymatteo/pi-extensions",
|
|
3
|
-
"version": "0.1.
|
|
3
|
+
"version": "0.1.2",
|
|
4
4
|
"type": "module",
|
|
5
5
|
"license": "AGPL-3.0-only",
|
|
6
6
|
"description": "Independently installable Pi extensions",
|
|
@@ -14,7 +14,9 @@
|
|
|
14
14
|
"pack:check": "node scripts/check-pack.mjs",
|
|
15
15
|
"dev": "node scripts/dev.mjs",
|
|
16
16
|
"smoke": "node scripts/smoke.mjs",
|
|
17
|
-
"context": "node scripts/context-audit.mjs"
|
|
17
|
+
"context": "node scripts/context-audit.mjs",
|
|
18
|
+
"screenshots": "node scripts/screenshots/shots.mjs",
|
|
19
|
+
"demo": "node scripts/screenshots/demo.mjs"
|
|
18
20
|
},
|
|
19
21
|
"devDependencies": {
|
|
20
22
|
"@earendil-works/pi-coding-agent": "1.4.0",
|
|
@@ -27,6 +29,7 @@
|
|
|
27
29
|
"node": ">=24"
|
|
28
30
|
},
|
|
29
31
|
"pi": {
|
|
32
|
+
"image": "https://raw.githubusercontent.com/saifymatteo/pi-extensions/main/demo.gif",
|
|
30
33
|
"extensions": [
|
|
31
34
|
"./packages/pi-load-pi-dotenv/src/index.ts",
|
|
32
35
|
"./packages/pi-bg-bash/src/index.ts",
|
|
@@ -68,6 +71,7 @@
|
|
|
68
71
|
],
|
|
69
72
|
"files": [
|
|
70
73
|
"packages",
|
|
74
|
+
"!packages/*/screenshot.png",
|
|
71
75
|
"inventory.json",
|
|
72
76
|
"README.md",
|
|
73
77
|
"NOTICE.md",
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# @saifymatteo/pi-bg-bash
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*The agent starts `npm test` with `background: true` and keeps working. When the job exits, its result arrives as a separate row. Ctrl+O expands that row to the output tail.*
|
|
6
|
+
|
|
3
7
|
## Purpose and registrations
|
|
4
8
|
|
|
5
9
|
Overrides `bash` and, when provided by the host, `powershell`, adding `background` and `job`. Registers `/bg`, result/call renderers, timeout injection, sleep-wait guard, session/settlement hooks, and shell provider discovery.
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# Claude subscription usage
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*An extra footer line shows Claude 5-hour and 7-day usage and when each window resets. Its colour changes at 75% and 90%.*
|
|
6
|
+
|
|
3
7
|
Displays Claude subscription rate-limit windows from provider message diagnostics.
|
|
4
8
|
|
|
5
9
|
## Try locally
|
|
@@ -14,7 +18,7 @@ This uses your normal personal resources unless you isolate the agent home. Foll
|
|
|
14
18
|
|
|
15
19
|
## Configuration and requirements
|
|
16
20
|
|
|
17
|
-
Uses rate-limit diagnostics from the external pi-claude-code provider; it does not poll Anthropic. Caches the most recent windows in <agent-dir>/state/claude-usage.json. No additional configuration.
|
|
21
|
+
Uses rate-limit diagnostics from the external pi-claude-code provider; it does not poll Anthropic. Caches the most recent windows in <agent-dir>/state/claude-usage.json. The footer is shown only while the selected model uses the `claude-code` provider, not for Claude models accessed through other providers. Switching providers hides or restores it immediately. No additional configuration.
|
|
18
22
|
|
|
19
23
|
## Compatibility and safety
|
|
20
24
|
|
|
@@ -77,7 +77,11 @@ export default function (pi: ExtensionAPI) {
|
|
|
77
77
|
let tick: ReturnType<typeof setInterval> | undefined;
|
|
78
78
|
|
|
79
79
|
function render() {
|
|
80
|
-
if (!ctx
|
|
80
|
+
if (!ctx) return;
|
|
81
|
+
if (ctx.model?.provider !== "claude-code" || !usage) {
|
|
82
|
+
ctx.ui.setStatus(STATUS_KEY, undefined);
|
|
83
|
+
return;
|
|
84
|
+
}
|
|
81
85
|
const now = Date.now();
|
|
82
86
|
const text = formatFooter(usage, now);
|
|
83
87
|
const p = peak(usage, now);
|
|
@@ -115,6 +119,11 @@ export default function (pi: ExtensionAPI) {
|
|
|
115
119
|
}
|
|
116
120
|
});
|
|
117
121
|
|
|
122
|
+
pi.on("model_select", async (_event, sessionCtx) => {
|
|
123
|
+
ctx = sessionCtx;
|
|
124
|
+
render();
|
|
125
|
+
});
|
|
126
|
+
|
|
118
127
|
pi.on("session_shutdown", async () => {
|
|
119
128
|
if (tick) clearInterval(tick);
|
|
120
129
|
tick = undefined;
|
|
@@ -0,0 +1,71 @@
|
|
|
1
|
+
import test from 'node:test';
|
|
2
|
+
import assert from 'node:assert/strict';
|
|
3
|
+
import { loadExtension } from '../../../test/load-extension.mjs';
|
|
4
|
+
|
|
5
|
+
const cached = { fiveHour: { pct: 42, resetsAt: 4_000_000_000_000 } };
|
|
6
|
+
|
|
7
|
+
function harness(t, initialModel, cache = cached) {
|
|
8
|
+
const hooks = {}, statuses = new Map(), timers = new Set(), writes = [];
|
|
9
|
+
const extension = loadExtension(new URL('../src/index.ts', import.meta.url), {
|
|
10
|
+
'node:fs': {
|
|
11
|
+
readFileSync: () => {
|
|
12
|
+
if (!cache) throw new Error('no cache');
|
|
13
|
+
return JSON.stringify(cache);
|
|
14
|
+
},
|
|
15
|
+
mkdirSync: () => {},
|
|
16
|
+
writeFileSync: (_path, contents) => writes.push(JSON.parse(contents)),
|
|
17
|
+
},
|
|
18
|
+
}, {
|
|
19
|
+
setInterval: fn => { timers.add(fn); return fn; },
|
|
20
|
+
clearInterval: fn => timers.delete(fn),
|
|
21
|
+
});
|
|
22
|
+
extension.default({ on: (name, fn) => { hooks[name] = fn; } });
|
|
23
|
+
const ui = {
|
|
24
|
+
setStatus: (key, value) => { if (value === undefined) statuses.delete(key); else statuses.set(key, value); },
|
|
25
|
+
theme: { fg: (_color, value) => value },
|
|
26
|
+
};
|
|
27
|
+
const context = model => ({ model, ui });
|
|
28
|
+
const ctx = context(initialModel);
|
|
29
|
+
t.after(() => hooks.session_shutdown());
|
|
30
|
+
return {
|
|
31
|
+
hooks, ctx, statuses, timers, writes,
|
|
32
|
+
select: model => hooks.model_select({ model }, context(model)),
|
|
33
|
+
};
|
|
34
|
+
}
|
|
35
|
+
|
|
36
|
+
for (const model of [undefined, { provider: 'openai', id: 'gpt-5' }, { provider: 'anthropic', id: 'claude-sonnet-4-6' }]) {
|
|
37
|
+
test(`cached Claude usage stays hidden for ${model?.provider ?? 'no model'}`, async t => {
|
|
38
|
+
const h = harness(t, model);
|
|
39
|
+
await h.hooks.session_start({}, h.ctx);
|
|
40
|
+
assert.equal(h.statuses.has('claude-usage'), false);
|
|
41
|
+
});
|
|
42
|
+
}
|
|
43
|
+
|
|
44
|
+
test('model changes hide and restore cached usage immediately, including timer refreshes', async t => {
|
|
45
|
+
const h = harness(t, { provider: 'claude-code', id: 'sonnet' });
|
|
46
|
+
await h.hooks.session_start({}, h.ctx);
|
|
47
|
+
assert.match(h.statuses.get('claude-usage'), /^Claude 5h 42%/);
|
|
48
|
+
|
|
49
|
+
await h.select({ provider: 'openai', id: 'gpt-5' });
|
|
50
|
+
assert.equal(h.statuses.has('claude-usage'), false);
|
|
51
|
+
for (const refresh of h.timers) refresh();
|
|
52
|
+
assert.equal(h.statuses.has('claude-usage'), false);
|
|
53
|
+
|
|
54
|
+
await h.select({ provider: 'claude-code', id: 'sonnet' });
|
|
55
|
+
assert.match(h.statuses.get('claude-usage'), /^Claude 5h 42%/);
|
|
56
|
+
});
|
|
57
|
+
|
|
58
|
+
test('diagnostics update the cache without exposing usage on another provider', async t => {
|
|
59
|
+
const h = harness(t, { provider: 'openai', id: 'gpt-5' }, null);
|
|
60
|
+
await h.hooks.session_start({}, h.ctx);
|
|
61
|
+
await h.hooks.message_end({ message: {
|
|
62
|
+
role: 'assistant',
|
|
63
|
+
diagnostics: [{ type: 'claude-code.rate-limit', details: {
|
|
64
|
+
unifiedWindows: { five_hour: { utilization: 0.75, resetsAt: 4_000_000_000 } },
|
|
65
|
+
} }],
|
|
66
|
+
} }, h.ctx);
|
|
67
|
+
assert.equal(h.statuses.has('claude-usage'), false);
|
|
68
|
+
assert.equal(h.writes.at(-1).fiveHour.pct, 75);
|
|
69
|
+
await h.select({ provider: 'claude-code', id: 'sonnet' });
|
|
70
|
+
assert.match(h.statuses.get('claude-usage'), /^Claude 5h 75%/);
|
|
71
|
+
});
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# Clipboard image attachment
|
|
2
2
|
|
|
3
|
+
![Alt+V pastes a clipboard image into the draft as [Image #1]](screenshot.png)
|
|
4
|
+
|
|
5
|
+
*Alt+V adds the clipboard image to the draft as `[Image #1]`. Sending the message sends the image to the model with it.*
|
|
6
|
+
|
|
3
7
|
Attaches clipboard images to the draft, including copied image files on Windows.
|
|
4
8
|
|
|
5
9
|
## Try locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# pi-codeblock-borders
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*The same reply, rendered by stock pi and then with this extension. Code fences become full-width rules, and `title=`, `:line-numbers` and `{2}` line focus are rendered.*
|
|
6
|
+
|
|
3
7
|
Render markdown code fences as full-width rules spanning the content area, and upgrade the fence info string with language decoration, titles, and line numbers that Pi's built-in renderer does not support.
|
|
4
8
|
|
|
5
9
|
## Running locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# @saifymatteo/pi-codegraph
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*The agent calls `codegraph_explore` and gets back the relevant source with its blast radius (expanded here with Ctrl+O). The footer shows the index status, which is synced after every turn.*
|
|
6
|
+
|
|
3
7
|
## Purpose and registrations
|
|
4
8
|
|
|
5
9
|
Registers `codegraph_explore`, `/codegraph-init`, `/codegraph-sync`, `/codegraph-status`, `/codegraph-unlock`, session startup/shutdown and settle-sync hooks, and footer status.
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# Compact tool cards
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*Four tool calls in a row fold into one summary card. Ctrl+O expands it, and each call keeps its own status and output.*
|
|
6
|
+
|
|
3
7
|
Groups consecutive tool calls into compact expandable transcript cards.
|
|
4
8
|
|
|
5
9
|
## Try locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# @saifymatteo/pi-decisions
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*The agent calls `classify` against a configured Decisions endpoint and gets structured answers with per-label probabilities.*
|
|
6
|
+
|
|
3
7
|
## Purpose and registrations
|
|
4
8
|
|
|
5
9
|
When configured, registers classifier providers and the `classify` tool (choice/bool/score; JSON state and optional PNG/JPEG/WebP images). No slash commands. Requires Pi's classifier model registry APIs (`findOfType`, `classify`, provider `classifiers`); older hosts may not implement them.
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# Pi dotenv loader
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*With this extension, variables from `~/.pi/agent/.env`, including `${VAR}` expansion, reach the tools the agent runs. Stock pi passes nothing through.*
|
|
6
|
+
|
|
3
7
|
Loads environment variables from the Pi agent directory without replacing existing environment values.
|
|
4
8
|
|
|
5
9
|
## Try locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# pi-model-go
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*`/model-go` lists OpenCode Go models sorted by effective price (input ÷ plan credit), with context size, intelligence score and credit multiplier.*
|
|
6
|
+
|
|
3
7
|
Cost-aware model picker showing registry prices/context, OpenCode Go credit multipliers and stencil.so intelligence scores. Source describes it as a port of rizal72/oc-models-pro.
|
|
4
8
|
|
|
5
9
|
## Running locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# @saifymatteo/pi-prestart-working
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*A slow `before_agent_start` hook normally leaves the UI silent after you press Enter. With this extension, the spinner names the extension that is running.*
|
|
6
|
+
|
|
3
7
|
## Purpose and registrations
|
|
4
8
|
|
|
5
9
|
Shows Pi's native "Working" spinner, labelled with the extension currently blocking (e.g. `⠙ Loading hindsight-coding-agents`), while `before_agent_start` hooks run. Pi only raises its working indicator at `turn_start`, so slow handlers otherwise leave the UI silent. No tools, commands, or configuration.
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# Agent configuration root guard
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*Stock pi lets the agent rewrite its own config. With the guard, writes into the Pi and Claude config roots are blocked, and the agent is told why.*
|
|
6
|
+
|
|
3
7
|
Blocks selected write/edit and simple bash writes into Pi and Claude configuration roots.
|
|
4
8
|
|
|
5
9
|
## Try locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# pi-session-finder
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*`/session-global` searches saved sessions across all projects. They are grouped by folder, with forks nested under their parent.*
|
|
6
|
+
|
|
3
7
|
Search saved sessions across projects, display their fork families, and open or fork a conversation into the current project.
|
|
4
8
|
|
|
5
9
|
## Running locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# pi-session-nav
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*Alt+O lists the conversation's messages. Pick one and press Enter to jump to it.*
|
|
6
|
+
|
|
3
7
|
Navigate visible user and assistant messages on the active conversation branch without adding transcript entries.
|
|
4
8
|
|
|
5
9
|
## Running locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# pi-session-rename
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*After the first prompt, the session gets a generated title. It appears in the footer and in the terminal tab title.*
|
|
6
|
+
|
|
3
7
|
Automatically title main interactive sessions from the first submitted prompt and optionally mirror the name to a Herdr tab. Manual names take precedence; failed generation retries on later prompts.
|
|
4
8
|
|
|
5
9
|
## Running locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# pi-skill-usage-log
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*`/skill-usage` shows which installed skills were loaded and which were actually used, across all projects.*
|
|
6
|
+
|
|
3
7
|
Passively record skill loads and subsequent tool activity, then show cross-project usage without adding the report to conversation context.
|
|
4
8
|
|
|
5
9
|
## Running locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# pi-tool-usage-log
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*`/tool-usage` totals every tool call: count, sessions, errors, nested calls, average duration and last use.*
|
|
6
|
+
|
|
3
7
|
Passively record direct and nested tool executions and show aggregate usage, errors, duration, selected model and thinking level.
|
|
4
8
|
|
|
5
9
|
## Running locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# @saifymatteo/pi-tts
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*The agent calls `text_to_speech`, and the WAV file from the configured speech endpoint is saved into the project.*
|
|
6
|
+
|
|
3
7
|
## Purpose and registrations
|
|
4
8
|
|
|
5
9
|
Registers `text_to_speech` when the registry has a `type: "tts"` model; no slash commands or playback. Sends narration to a service and saves a new WAV, never overwriting an existing output. Uses standard fetch/streams and Pi tool registration; requires filesystem hard-link support for atomic publication.
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# Inline user image previews
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*An attached image gets a collapsible preview row in the transcript. Ctrl+O opens it inline. Stock pi shows nothing.*
|
|
6
|
+
|
|
3
7
|
Adds collapsible UI-only previews for newly submitted image attachments.
|
|
4
8
|
|
|
5
9
|
## Try locally
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# @saifymatteo/pi-vision
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*The main model here is text-only. Stock pi drops the image. With pi-vision, a configured vision model describes the image and the main model works from that description.*
|
|
6
|
+
|
|
3
7
|
## Purpose and registrations
|
|
4
8
|
|
|
5
9
|
Registers a `context` hook, not a tool or slash command. Preserves native images for image-capable session models by default. For text-only/unknown models it replaces image blocks with model-generated descriptions. Optional `describe` / `describe-old` policies self-describe all/older images using the current model on OpenAI Chat Completions connections.
|
|
@@ -12,7 +16,7 @@ Resolution also consults agent `models.json`, `models-store.json`, `auth.json`,
|
|
|
12
16
|
|
|
13
17
|
## Bundled skill
|
|
14
18
|
|
|
15
|
-
The package bundles the `vision-tools` skill (`skills/vision-tools`): `bin/` launcher CLIs (`crop`, `detect`, `glance`, `ground`, `trace`, `vision-doctor`), Python tools under `scripts/`, reference guides, and `SKILL.md`. The extension's guidance messages point the agent at `vision-doctor` and the skill. Requires Python 3 with Pillow (PIL); the CLIs are Python scripts, not compiled binaries. Based on [Anionex/agent-vision-toolkit](https://github.com/Anionex/agent-vision-toolkit) (MIT), modified in this repository — see [NOTICE.md](NOTICE.md).
|
|
19
|
+
The package bundles the `vision-tools` skill (`skills/vision-tools`): `bin/` launcher CLIs (`crop`, `detect`, `glance`, `ground`, `trace`, `vision-doctor`), Python tools under `scripts/`, reference guides, and `SKILL.md`. Invoke the launchers with Python and their absolute skill-relative paths, for example `python3 "<skill-directory>/bin/crop" image.png --region 0,0,100,100 -o detail.png`. Follow the interpreter-selection guidance in `SKILL.md` if Windows exposes Python as `python` or `py -3`. No PATH registration, executable bit, or agent-home shim is required. The extension does not modify PATH. Automated long-screenshot OCR prefers the bundled `bin/glance` with the current Python interpreter, falling back to PATH only if that script is missing. The extension's guidance messages point the agent at `vision-doctor` and the skill. Requires Python 3 with Pillow (PIL); the CLIs are Python scripts, not compiled binaries. Based on [Anionex/agent-vision-toolkit](https://github.com/Anionex/agent-vision-toolkit) (MIT), modified in this repository — see [NOTICE.md](NOTICE.md).
|
|
16
20
|
|
|
17
21
|
## Host limitations and security
|
|
18
22
|
|
|
@@ -33,20 +33,35 @@ For browser or accessible GUI work, prefer DOM/accessibility information for
|
|
|
33
33
|
controls and state. Use screenshot coordinates when that information is
|
|
34
34
|
unavailable or the task specifically concerns appearance.
|
|
35
35
|
|
|
36
|
+
## Run the packaged scripts
|
|
37
|
+
|
|
38
|
+
1. Resolve `<skill-dir>` to the absolute directory containing this `SKILL.md`,
|
|
39
|
+
using the skill location supplied by the host. Replace that placeholder in
|
|
40
|
+
commands below; keep script paths quoted and keep the project's working
|
|
41
|
+
directory.
|
|
42
|
+
2. Select a Python 3 interpreter: try `python3 --version`; on Windows, use
|
|
43
|
+
`python --version` or `py -3 --version` if needed. Substitute that command
|
|
44
|
+
for `python3` in the examples.
|
|
45
|
+
3. Invoke `bin/<tool>` and `scripts/<name>.py` through that interpreter. The
|
|
46
|
+
extensionless files in `bin/` are Python scripts, not installed binaries.
|
|
47
|
+
This works without PATH registration, executable permissions, or agent-home
|
|
48
|
+
symlinks. Short tool names in prose refer to these same script paths.
|
|
49
|
+
|
|
50
|
+
Use `--help` on the selected script for additional options.
|
|
51
|
+
|
|
36
52
|
## Local tools: use when their output is needed
|
|
37
53
|
|
|
38
|
-
|
|
39
|
-
|
|
40
|
-
load resources referenced by the page.
|
|
54
|
+
These tools do not use the vision API; HTML rendering can load resources
|
|
55
|
+
referenced by the page.
|
|
41
56
|
|
|
42
57
|
| Need | Command |
|
|
43
58
|
| --- | --- |
|
|
44
|
-
| Reusable crop or close-up | `crop shot.png --region X1,Y1,X2,Y2 -o detail.png` |
|
|
45
|
-
| Locate pixel differences | `python3 scripts/pixel_diff.py before.png after.png` |
|
|
46
|
-
| Measure a region's palette | `python3 scripts/dominant_colors.py shot.png --region X1,Y1,X2,Y2` |
|
|
47
|
-
| Vectorize a flat graphic | `trace icon.png -o icon.svg` |
|
|
48
|
-
| Extract a transparent foreground | `python3 scripts/extract_fg.py shot.png --region X1,Y1,X2,Y2 -o icon.png` |
|
|
49
|
-
| Render an HTML file | `python3 scripts/html_shot.py page.html --width 1440 --height 900 -o page.png` |
|
|
59
|
+
| Reusable crop or close-up | `python3 "<skill-dir>/bin/crop" shot.png --region X1,Y1,X2,Y2 -o detail.png` |
|
|
60
|
+
| Locate pixel differences | `python3 "<skill-dir>/scripts/pixel_diff.py" before.png after.png` |
|
|
61
|
+
| Measure a region's palette | `python3 "<skill-dir>/scripts/dominant_colors.py" shot.png --region X1,Y1,X2,Y2` |
|
|
62
|
+
| Vectorize a flat graphic | `python3 "<skill-dir>/bin/trace" icon.png -o icon.svg` |
|
|
63
|
+
| Extract a transparent foreground | `python3 "<skill-dir>/scripts/extract_fg.py" shot.png --region X1,Y1,X2,Y2 -o icon.png` |
|
|
64
|
+
| Render an HTML file | `python3 "<skill-dir>/scripts/html_shot.py" page.html --width 1440 --height 900 -o page.png` |
|
|
50
65
|
|
|
51
66
|
- `crop --scale 4` enlarges a crop; it adds no missing detail. Coordinates
|
|
52
67
|
measured on that output use the enlarged grid, not the original image.
|
|
@@ -84,11 +99,11 @@ separately configured vision chain, not to the main agent model automatically.
|
|
|
84
99
|
Structured output does not guarantee better recognition or localization.
|
|
85
100
|
|
|
86
101
|
```bash
|
|
87
|
-
glance shot.png -q "Read the small status label."
|
|
88
|
-
glance shot.png --region X1,Y1,X2,Y2 --ocr
|
|
89
|
-
glance before.png after.png -q "What changed semantically?"
|
|
90
|
-
ground shot.png "the send button"
|
|
91
|
-
detect shot.png "buttons"
|
|
102
|
+
python3 "<skill-dir>/bin/glance" shot.png -q "Read the small status label."
|
|
103
|
+
python3 "<skill-dir>/bin/glance" shot.png --region X1,Y1,X2,Y2 --ocr
|
|
104
|
+
python3 "<skill-dir>/bin/glance" before.png after.png -q "What changed semantically?"
|
|
105
|
+
python3 "<skill-dir>/bin/ground" shot.png "the send button"
|
|
106
|
+
python3 "<skill-dir>/bin/detect" shot.png "buttons"
|
|
92
107
|
```
|
|
93
108
|
|
|
94
109
|
`ground` and `detect` return estimated boxes in original-image pixels, including
|
|
@@ -98,9 +113,9 @@ For multi-image questions, pass the images in one call so the model sees them
|
|
|
98
113
|
together. Native side-by-side inspection is also sufficient for semantic
|
|
99
114
|
comparison; use pixel diff when numerical differences matter.
|
|
100
115
|
|
|
101
|
-
The long-screenshot OCR script
|
|
102
|
-
|
|
103
|
-
|
|
116
|
+
The long-screenshot OCR script invokes this skill's `bin/glance` with its
|
|
117
|
+
current Python interpreter, falling back to PATH only if the bundled script
|
|
118
|
+
is missing. Its `--split-only` mode makes chunks without API requests.
|
|
104
119
|
|
|
105
120
|
### Configuration and failures
|
|
106
121
|
|
|
@@ -112,10 +127,11 @@ conversion on supported connections. Text-only or unknown-capability models
|
|
|
112
127
|
still use the configured compatibility bridge. This policy does not route CLI
|
|
113
128
|
calls through the main model.
|
|
114
129
|
|
|
115
|
-
On a model-backed CLI configuration/API error, run
|
|
116
|
-
`vision-doctor
|
|
117
|
-
a test image. Relay failures honestly.
|
|
118
|
-
available; report if the scripted output
|
|
130
|
+
On a model-backed CLI configuration/API error, run
|
|
131
|
+
`python3 "<skill-dir>/bin/vision-doctor"`. Add `--smoke` only when a real
|
|
132
|
+
diagnostic request is needed; it sends a test image. Relay failures honestly.
|
|
133
|
+
Continue with native inspection when available; report if the scripted output
|
|
134
|
+
cannot be produced. Keep provider
|
|
119
135
|
changes user-controlled: do not scrape credentials, hand-export secret overrides,
|
|
120
136
|
or silently switch to an unconfigured provider.
|
|
121
137
|
|
|
@@ -12,7 +12,7 @@ beyond a temporary file's lifetime. Inspect the script's help before choosing
|
|
|
12
12
|
options:
|
|
13
13
|
|
|
14
14
|
```bash
|
|
15
|
-
python3 scripts/long_screenshot_ocr.py --help
|
|
15
|
+
python3 "<skill-dir>/scripts/long_screenshot_ocr.py" --help
|
|
16
16
|
```
|
|
17
17
|
|
|
18
18
|
Prefer native chunk reading by the main agent; the glance-backed script is
|
|
@@ -22,7 +22,7 @@ splitter; verify joins even when the cut falls in a low-content band.
|
|
|
22
22
|
### Native chunk reading
|
|
23
23
|
|
|
24
24
|
```bash
|
|
25
|
-
python3 scripts/long_screenshot_ocr.py work/page.png --split-only
|
|
25
|
+
python3 "<skill-dir>/scripts/long_screenshot_ocr.py" work/page.png --split-only
|
|
26
26
|
```
|
|
27
27
|
|
|
28
28
|
This writes ordered chunk images and `manifest.json` without vision calls.
|
|
@@ -38,10 +38,10 @@ Choose the content mode; this path makes configured vision calls:
|
|
|
38
38
|
|
|
39
39
|
```bash
|
|
40
40
|
# Web pages, documents, logs, tables, and other general content
|
|
41
|
-
python3 scripts/long_screenshot_ocr.py work/page.png -o work/page.ocr.md
|
|
41
|
+
python3 "<skill-dir>/scripts/long_screenshot_ocr.py" work/page.png -o work/page.ocr.md
|
|
42
42
|
|
|
43
43
|
# Chat histories: preserve message grouping, senders, timestamps, and quotes
|
|
44
|
-
python3 scripts/long_screenshot_ocr.py work/chat.png --mode chat -o work/chat.ocr.md
|
|
44
|
+
python3 "<skill-dir>/scripts/long_screenshot_ocr.py" work/chat.png --mode chat -o work/chat.ocr.md
|
|
45
45
|
```
|
|
46
46
|
|
|
47
47
|
Without `-o`, write `<input-stem>.ocr.md`; store chunks, sidecars, the manifest,
|
|
@@ -63,7 +63,7 @@ Use `--resume` after an interrupted run. A chunk is reused only when its image,
|
|
|
63
63
|
mode, and custom prompt fingerprint still match:
|
|
64
64
|
|
|
65
65
|
```bash
|
|
66
|
-
python3 scripts/long_screenshot_ocr.py work/chat.png --mode chat --resume -o work/chat.ocr.md
|
|
66
|
+
python3 "<skill-dir>/scripts/long_screenshot_ocr.py" work/chat.png --mode chat --resume -o work/chat.ocr.md
|
|
67
67
|
```
|
|
68
68
|
|
|
69
69
|
If the defaults produce awkward chunks, rerun with `--target-height`,
|
|
@@ -85,8 +85,8 @@ context; do not make tiny OCR tiles unless the source text is unusually small.
|
|
|
85
85
|
unresolved, targeted specialist OCR is available:
|
|
86
86
|
|
|
87
87
|
```bash
|
|
88
|
-
glance work/page_chunks/chunk_002.png --ocr "Re-check the final five lines carefully."
|
|
89
|
-
glance work/page_chunks/chunk_002.png --region X1,Y1,X2,Y2 --ocr
|
|
88
|
+
python3 "<skill-dir>/bin/glance" work/page_chunks/chunk_002.png --ocr "Re-check the final five lines carefully."
|
|
89
|
+
python3 "<skill-dir>/bin/glance" work/page_chunks/chunk_002.png --region X1,Y1,X2,Y2 --ocr
|
|
90
90
|
```
|
|
91
91
|
|
|
92
92
|
5. Keep visible spelling and punctuation verbatim. Write `[unreadable]` for
|
|
@@ -25,16 +25,16 @@ pixels as foreground.
|
|
|
25
25
|
|
|
26
26
|
```bash
|
|
27
27
|
# Known tight region
|
|
28
|
-
python3 scripts/extract_fg.py shot.png \
|
|
28
|
+
python3 "<skill-dir>/scripts/extract_fg.py" shot.png \
|
|
29
29
|
--region X1,Y1,X2,Y2 -o asset.png
|
|
30
30
|
|
|
31
31
|
# Dark or gray line art
|
|
32
|
-
python3 scripts/extract_fg.py shot.png \
|
|
32
|
+
python3 "<skill-dir>/scripts/extract_fg.py" shot.png \
|
|
33
33
|
--region X1,Y1,X2,Y2 --mode dark -o asset.png
|
|
34
34
|
|
|
35
35
|
# Center and enlarge a small visual before automatic extraction
|
|
36
|
-
crop shot.png --region X1,Y1,X2,Y2 --scale 4 -o work/asset4x.png
|
|
37
|
-
python3 scripts/extract_fg.py work/asset4x.png
|
|
36
|
+
python3 "<skill-dir>/bin/crop" shot.png --region X1,Y1,X2,Y2 --scale 4 -o work/asset4x.png
|
|
37
|
+
python3 "<skill-dir>/scripts/extract_fg.py" work/asset4x.png
|
|
38
38
|
```
|
|
39
39
|
|
|
40
40
|
Use `--exclude-color '#RRGGBB'` when a connected background color would
|
|
@@ -47,9 +47,9 @@ clean alpha edges, and no neighboring fragments before reuse.
|
|
|
47
47
|
## Rebuild an SVG only when the deliverable needs one
|
|
48
48
|
|
|
49
49
|
```bash
|
|
50
|
-
trace asset.png -o asset.svg
|
|
51
|
-
trace asset.png --polygon -o asset.svg
|
|
52
|
-
trace shot.png --region X1,Y1,X2,Y2 -o asset.svg
|
|
50
|
+
python3 "<skill-dir>/bin/trace" asset.png -o asset.svg
|
|
51
|
+
python3 "<skill-dir>/bin/trace" asset.png --polygon -o asset.svg
|
|
52
|
+
python3 "<skill-dir>/bin/trace" shot.png --region X1,Y1,X2,Y2 -o asset.svg
|
|
53
53
|
```
|
|
54
54
|
|
|
55
55
|
Ship traced paths directly for organic or irregular shapes. For rectangles,
|
|
@@ -140,13 +140,13 @@ extraction, tighten the final region to the visual's own ink; adjacent text or
|
|
|
140
140
|
rules inside the region can become foreground components too.
|
|
141
141
|
|
|
142
142
|
```bash
|
|
143
|
-
crop shot.png --region X1,Y1,X2,Y2 --scale 4 -o work/icon4x.png
|
|
144
|
-
python3 scripts/extract_fg.py work/icon4x.png
|
|
143
|
+
python3 "<skill-dir>/bin/crop" shot.png --region X1,Y1,X2,Y2 --scale 4 -o work/icon4x.png
|
|
144
|
+
python3 "<skill-dir>/scripts/extract_fg.py" work/icon4x.png
|
|
145
145
|
|
|
146
146
|
# Or extract directly from a known tight region:
|
|
147
|
-
python3 scripts/extract_fg.py shot.png \
|
|
147
|
+
python3 "<skill-dir>/scripts/extract_fg.py" shot.png \
|
|
148
148
|
--region X1,Y1,X2,Y2 -o work/icon.png
|
|
149
|
-
python3 scripts/extract_fg.py shot.png \
|
|
149
|
+
python3 "<skill-dir>/scripts/extract_fg.py" shot.png \
|
|
150
150
|
--region X1,Y1,X2,Y2 --mode dark -o work/logo.png
|
|
151
151
|
```
|
|
152
152
|
|
|
@@ -456,13 +456,13 @@ def command_for_path(path: Path) -> list[str]:
|
|
|
456
456
|
|
|
457
457
|
|
|
458
458
|
def resolve_glance_command() -> list[str]:
|
|
459
|
+
bundled_glance = Path(__file__).resolve().parents[1] / "bin" / "glance"
|
|
460
|
+
if bundled_glance.is_file():
|
|
461
|
+
return [sys.executable, str(bundled_glance)]
|
|
459
462
|
discovered = shutil.which("glance")
|
|
460
463
|
if discovered:
|
|
461
464
|
return command_for_path(Path(discovered))
|
|
462
|
-
|
|
463
|
-
if repository_glance.is_file():
|
|
464
|
-
return command_for_path(repository_glance)
|
|
465
|
-
raise FileNotFoundError("glance is not on PATH; install the toolkit first")
|
|
465
|
+
raise FileNotFoundError("glance is not in this skill directory or on PATH")
|
|
466
466
|
|
|
467
467
|
|
|
468
468
|
def ocr_prompt(mode: str, index: int, total: int, custom: str | None) -> str:
|
|
@@ -0,0 +1,71 @@
|
|
|
1
|
+
import test from 'node:test';
|
|
2
|
+
import assert from 'node:assert/strict';
|
|
3
|
+
import { readFileSync, readdirSync, existsSync } from 'node:fs';
|
|
4
|
+
import { join } from 'node:path';
|
|
5
|
+
import { fileURLToPath } from 'node:url';
|
|
6
|
+
import { spawnSync } from 'node:child_process';
|
|
7
|
+
|
|
8
|
+
const skill = fileURLToPath(new URL('../skills/vision-tools/', import.meta.url));
|
|
9
|
+
|
|
10
|
+
test('skill command examples invoke bundled Python scripts independently of PATH and cwd', () => {
|
|
11
|
+
const docs = [join(skill, 'SKILL.md'), ...readdirSync(join(skill, 'references')).map(name => join(skill, 'references', name))];
|
|
12
|
+
const launchers = new Set();
|
|
13
|
+
for (const doc of docs) {
|
|
14
|
+
const source = readFileSync(doc, 'utf8');
|
|
15
|
+
for (const [, commands] of source.matchAll(/```bash\n([\s\S]*?)```/g)) {
|
|
16
|
+
assert.equal(/^\s*(?:crop|detect|glance|ground|trace|vision-doctor)\s+\S/m.test(commands), false, `${doc}: bare launcher command`);
|
|
17
|
+
}
|
|
18
|
+
assert.equal(/python3 scripts\//.test(source), false, `${doc}: cwd-relative script path`);
|
|
19
|
+
for (const [, relative] of source.matchAll(/python3 "<skill-dir>\/([^"]+)"/g)) {
|
|
20
|
+
assert.ok(existsSync(join(skill, relative)), `${doc}: missing bundled script ${relative}`);
|
|
21
|
+
if (relative.startsWith('bin/')) launchers.add(relative.slice(4));
|
|
22
|
+
}
|
|
23
|
+
}
|
|
24
|
+
assert.deepEqual([...launchers].sort(), ['crop', 'detect', 'glance', 'ground', 'trace', 'vision-doctor'].sort());
|
|
25
|
+
});
|
|
26
|
+
|
|
27
|
+
test('automated OCR finds bundled glance without PATH, including non-executable paths with spaces', t => {
|
|
28
|
+
const python = [['python3'], ['python'], ['py', '-3']].find(([command, ...args]) => {
|
|
29
|
+
const result = spawnSync(command, [...args, '--version'], { encoding: 'utf8', timeout: 10_000 });
|
|
30
|
+
return result.status === 0 && /^Python 3\./.test(result.stdout);
|
|
31
|
+
});
|
|
32
|
+
if (!python) return t.skip('Python 3 is not installed');
|
|
33
|
+
const code = `
|
|
34
|
+
import importlib.util
|
|
35
|
+
from pathlib import Path
|
|
36
|
+
import subprocess
|
|
37
|
+
import sys
|
|
38
|
+
import tempfile
|
|
39
|
+
from unittest.mock import patch
|
|
40
|
+
|
|
41
|
+
source = Path(sys.argv[1])
|
|
42
|
+
spec = importlib.util.spec_from_file_location("long_screenshot_ocr", source)
|
|
43
|
+
module = importlib.util.module_from_spec(spec)
|
|
44
|
+
sys.modules[spec.name] = module
|
|
45
|
+
spec.loader.exec_module(module)
|
|
46
|
+
with tempfile.TemporaryDirectory(prefix="vision skill ") as directory:
|
|
47
|
+
skill = Path(directory)
|
|
48
|
+
script = skill / "scripts" / "long_screenshot_ocr.py"
|
|
49
|
+
script.parent.mkdir()
|
|
50
|
+
script.touch()
|
|
51
|
+
glance = skill / "bin" / "glance"
|
|
52
|
+
glance.parent.mkdir()
|
|
53
|
+
glance.write_text('print("bundled glance reached")', encoding="utf-8")
|
|
54
|
+
with patch.object(module, "__file__", str(script)), patch.object(module.shutil, "which", return_value=None):
|
|
55
|
+
command = module.resolve_glance_command()
|
|
56
|
+
assert command == [sys.executable, str(glance)], command
|
|
57
|
+
result = subprocess.run(command, capture_output=True, text=True, check=True)
|
|
58
|
+
assert result.stdout.strip() == "bundled glance reached", result.stdout
|
|
59
|
+
with patch.object(module, "__file__", str(script)), patch.object(module.shutil, "which", return_value="/unrelated/glance"):
|
|
60
|
+
assert module.resolve_glance_command() == command, "prefer the packaged script over unrelated PATH commands"
|
|
61
|
+
glance.unlink()
|
|
62
|
+
external = skill / "custom_glance.py"
|
|
63
|
+
external.touch()
|
|
64
|
+
with patch.object(module, "__file__", str(script)), patch.object(module.shutil, "which", return_value=str(external)):
|
|
65
|
+
assert module.resolve_glance_command() == [sys.executable, str(external)], "retain the external fallback"
|
|
66
|
+
`;
|
|
67
|
+
const result = spawnSync(python[0], [...python.slice(1), '-c', code, join(skill, 'scripts', 'long_screenshot_ocr.py')], {
|
|
68
|
+
encoding: 'utf8', timeout: 20_000, env: { ...process.env, PYTHONDONTWRITEBYTECODE: '1' },
|
|
69
|
+
});
|
|
70
|
+
assert.equal(result.status, 0, result.stderr || result.error?.message);
|
|
71
|
+
});
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# @saifymatteo/pi-wait
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*The agent starts a build in the background, then calls `wait` on `shell:1`. The job's output comes back in the same turn. Needs pi-bg-bash.*
|
|
6
|
+
|
|
3
7
|
## Purpose and registrations
|
|
4
8
|
|
|
5
9
|
Registers the `wait` tool, joining 1..64 unique explicit `shell:<id>` / `agent:<id>` jobs. Returns results in requested order. Timeout releases claims without stopping work; abort/session changes invalidate the join. No commands or configuration files; no external CLI/model service.
|
|
@@ -1,5 +1,9 @@
|
|
|
1
1
|
# @saifymatteo/pi-zvec
|
|
2
2
|
|
|
3
|
+
|
|
4
|
+
|
|
5
|
+
*The agent calls `zvec_search` in hybrid mode, which combines lexical and semantic search (expanded here with Ctrl+O). The footer shows the index status.*
|
|
6
|
+
|
|
3
7
|
## Purpose and registrations
|
|
4
8
|
|
|
5
9
|
Registers `zvec_search` (exact/fts/vector/hybrid/auto), `/zvec-init`, `/zvec-sync`, `/zvec-status`, session and settle-sync hooks, and footer status.
|