@withgauge/cli 0.21.0 → 0.22.0
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 +393 -75
- package/dist/batchWorker.js +2 -0
- package/dist/chunk-CIJ4U33Y.js +70 -0
- package/dist/chunk-FCK3MQPV.js +2 -0
- package/dist/chunk-KLWQOTEK.js +3 -0
- package/dist/chunk-PSBF4TEK.js +1 -0
- package/dist/chunk-RB7Q72OV.js +1 -0
- package/dist/index.js +561 -11697
- package/dist/telemetryWorker.js +1 -0
- package/dist/worker.js +1 -0
- package/package.json +13 -4
package/README.md
CHANGED
|
@@ -7,31 +7,220 @@ live from your terminal.
|
|
|
7
7
|
```sh
|
|
8
8
|
npm install -g @withgauge/cli
|
|
9
9
|
gauge update # update the CLI to the latest published version
|
|
10
|
-
gauge onboard # sign in
|
|
10
|
+
gauge onboard # sign in and create a workspace (account only)
|
|
11
11
|
gauge auth login # (or just sign in)
|
|
12
12
|
gauge orgs use <org> # set your default organization
|
|
13
|
+
gauge init # initialize a workspace; local is the default
|
|
14
|
+
gauge init --hosted # set a private hosted backend preference
|
|
15
|
+
gauge init --cases # optionally scaffold shared gauge-cases/
|
|
13
16
|
```
|
|
14
17
|
|
|
15
|
-
Requires Node.js ≥ 22.
|
|
18
|
+
Requires Node.js ≥ 22.13.
|
|
16
19
|
|
|
17
|
-
|
|
18
|
-
|
|
19
|
-
|
|
20
|
+
## Workspace and local storage
|
|
21
|
+
|
|
22
|
+
`gauge init --local` works without Git. It writes a minimal shared manifest:
|
|
23
|
+
|
|
24
|
+
```json
|
|
25
|
+
{ "version": 2, "inputs": {} }
|
|
26
|
+
```
|
|
27
|
+
|
|
28
|
+
The private `.gauge/config.json` contains `{ "version": 1, "backend": "local" }`.
|
|
29
|
+
`.gauge/.gitignore` ignores all workspace state. SQLite in `.gauge/gauge.sqlite`
|
|
30
|
+
stores saved measurements, revisions, launch/session summaries, and published
|
|
31
|
+
analyses. Raw transcripts, normalized events, diffs, and diagnostics remain under
|
|
32
|
+
`.gauge/runs/`; input blobs and batch lifecycle files remain alongside them.
|
|
33
|
+
|
|
34
|
+
Moving the workspace preserves identity. Fresh clones or linked worktrees start
|
|
35
|
+
independently; switching Git branches keeps local history. Do not treat `.gauge/`
|
|
36
|
+
as a disposable cache. Credentials and user preferences stay user-level.
|
|
37
|
+
`gauge run repair --local` rebuilds session summaries from durable artifacts while
|
|
38
|
+
preserving authored measurements and analyses. It does not rerun sessions or judges.
|
|
39
|
+
|
|
40
|
+
## Analyze local sessions
|
|
41
|
+
|
|
42
|
+
Your coding agent reads evidence and publishes a structured analysis:
|
|
43
|
+
|
|
44
|
+
```sh
|
|
45
|
+
gauge instructions session-analysis --local
|
|
46
|
+
gauge run analysis schema --local > analysis.schema.json
|
|
47
|
+
gauge run get <run-id> --logs --local -o json
|
|
48
|
+
# Author analysis.json against the printed schema and session evidence.
|
|
49
|
+
gauge run analysis save <run-id> --file analysis.json --local
|
|
50
|
+
gauge run insight <run-id> --local
|
|
51
|
+
```
|
|
52
|
+
|
|
53
|
+
Publication validates the versioned schema, criterion coverage, and evidence
|
|
54
|
+
references, then saves findings, product choices, and criterion judgments in one
|
|
55
|
+
transaction. Reanalysis creates a revision and updates the current analysis.
|
|
56
|
+
`run insight` leads with Friction, Objections, and Opportunities in either backend.
|
|
57
|
+
A Markdown file alone does not mark a session analyzed. Unknown future schema
|
|
58
|
+
versions produce an upgrade error without altering saved data. Database schema
|
|
59
|
+
migrations preserve analyses and never silently reinterpret old judgments.
|
|
60
|
+
|
|
61
|
+
## Local Docker runtime
|
|
62
|
+
|
|
63
|
+
Local runs need Docker running Linux containers and an agent credential. Linux
|
|
64
|
+
is the container OS, not a requirement for your host: Docker Desktop on macOS
|
|
65
|
+
provides a Linux daemon and shares the local files used by sessions.
|
|
66
|
+
Install the runtime without a Gauge account or a source checkout:
|
|
67
|
+
|
|
68
|
+
```sh
|
|
69
|
+
gauge local runtime pull
|
|
70
|
+
gauge local runtime inspect -o json
|
|
71
|
+
gauge run --local --dry-run
|
|
72
|
+
```
|
|
73
|
+
|
|
74
|
+
Each CLI release pins its tested public `ghcr.io/gauge-sh/gauge-runtime` image by
|
|
75
|
+
immutable digest. Inspection reports the reference, local image ID, registry
|
|
76
|
+
digest, contract, platform, and readiness errors. `ready` covers local image
|
|
77
|
+
metadata; agent credential checks happen separately. Neither inspection nor a
|
|
78
|
+
dry run downloads an image. `pull` explicitly downloads the supported image;
|
|
79
|
+
previously pulled images can be reused offline (agent calls still need network
|
|
80
|
+
access). Pull progress goes to stderr so JSON output remains parseable.
|
|
81
|
+
|
|
82
|
+
The runtime includes native `linux/amd64` and `linux/arm64` images. Docker selects
|
|
83
|
+
the matching image on Intel/AMD or Apple Silicon/ARM machines. No architecture
|
|
84
|
+
emulation is needed. Docker must be able to bind-mount this machine's temporary
|
|
85
|
+
and input directories; an unrelated remote Docker server cannot access them.
|
|
86
|
+
On Windows, run the CLI inside WSL because Gauge uses Unix sockets and UID/GID.
|
|
87
|
+
|
|
88
|
+
## Local agent credentials
|
|
89
|
+
|
|
90
|
+
Local provider credentials are separate from a Gauge account and hosted credential
|
|
91
|
+
brokering. Use native harness logins or configure a named environment reference:
|
|
92
|
+
|
|
93
|
+
```sh
|
|
94
|
+
gauge local profiles set work --agent CODEX_CLI --provider openai --auth api_key --env TEAM_OPENAI_KEY
|
|
95
|
+
gauge local profiles use work
|
|
96
|
+
gauge local profiles inspect work -o json
|
|
97
|
+
gauge run --local --profile work --dry-run -o json
|
|
98
|
+
```
|
|
99
|
+
|
|
100
|
+
`--env` takes the variable's name, never its secret value. Profiles store only
|
|
101
|
+
references; list and inspect output do not expose credentials. Update a profile
|
|
102
|
+
with `set ... --replace` or remove it with `remove <profile>` without deleting
|
|
103
|
+
the native login. Local readiness checks do not verify provider acceptance or
|
|
104
|
+
model access. Use `gauge local profiles --help` for profile commands and
|
|
105
|
+
`gauge instructions --local` for the local workflow.
|
|
106
|
+
|
|
107
|
+
## Anonymous CLI telemetry
|
|
108
|
+
|
|
109
|
+
The installed CLI collects command usage by default to help improve Gauge.
|
|
110
|
+
Opt out for your user account or for a single process:
|
|
111
|
+
|
|
112
|
+
```sh
|
|
113
|
+
gauge telemetry disable
|
|
114
|
+
gauge telemetry status
|
|
115
|
+
gauge telemetry enable
|
|
116
|
+
GAUGE_TELEMETRY_DISABLED=1 gauge run --verify
|
|
117
|
+
```
|
|
118
|
+
|
|
119
|
+
The environment variable overrides the saved preference. These controls work
|
|
120
|
+
without an account or network access and support `-o json`. They apply to CLI
|
|
121
|
+
command telemetry; authenticated Gauge API requests have separate server-side
|
|
122
|
+
usage analytics.
|
|
123
|
+
|
|
124
|
+
Each `cli_command_invoked` event contains only the canonical command name
|
|
125
|
+
(for example, `run list`), its UTC invocation timestamp, CLI version, and a
|
|
126
|
+
random installation UUID. Arguments, flag values, file paths, prompts, file
|
|
127
|
+
contents, environment variables, errors, and account identifiers are excluded.
|
|
128
|
+
The UUID is independent of login and is never linked to your Gauge account.
|
|
129
|
+
This is pseudonymous usage tracking: an installation can be recognized across
|
|
130
|
+
commands, but no personal identity is sent. HTTP requests necessarily expose a
|
|
131
|
+
source IP to the receiving infrastructure; event properties mask IP addresses,
|
|
132
|
+
disable GeoIP enrichment, and disable person-profile creation.
|
|
133
|
+
|
|
134
|
+
Preferences live in `~/.config/gauge/telemetry.json` and the UUID in
|
|
135
|
+
`~/.config/gauge/telemetry-id` (respecting `XDG_CONFIG_HOME`). They are shared
|
|
136
|
+
across repositories, independently of each repository's `.gauge/` results.
|
|
137
|
+
Deleting `telemetry-id` resets the identity on the next collected command.
|
|
138
|
+
Consumers are installations, not necessarily individual people. Session
|
|
139
|
+
analysis can group an installation's events using a 30-minute inactivity gap;
|
|
140
|
+
the CLI does not generate a different identity for each invocation.
|
|
141
|
+
|
|
142
|
+
Collection occurs when a parsed command is about to run, including commands
|
|
143
|
+
whose handlers subsequently fail. Help, version, parsing errors, telemetry
|
|
144
|
+
controls, and source development runs are excluded. A detached worker handles
|
|
145
|
+
all telemetry storage and network access, with a two-second request timeout
|
|
146
|
+
and a three-second worker lifetime. Commands never wait for delivery, and
|
|
147
|
+
failures are silent. There are no retries or offline queue, so events may be
|
|
148
|
+
lost. Corrupt or unreadable preferences suppress collection until repaired.
|
|
149
|
+
|
|
150
|
+
Events go to Gauge's existing PostHog project using its public ingestion key.
|
|
151
|
+
The [capture API](https://posthog.com/docs/api/capture) and
|
|
152
|
+
[anonymous event controls](https://posthog.com/docs/product-analytics/capture-events#advanced-anonymous-vs-identified-events)
|
|
153
|
+
describe the transport and person-profile behavior.
|
|
154
|
+
|
|
155
|
+
## Local or hosted
|
|
156
|
+
|
|
157
|
+
For an end-to-end improvement loop, run `gauge instructions hillclimb`. It guides
|
|
158
|
+
your coding agent to identify product inputs, propose a representative pilot,
|
|
159
|
+
inspect and publish findings, implement a focused change, and rerun to compare.
|
|
160
|
+
Init and the generated `AGENTS.md` point to this guidance. `gauge install-skill`
|
|
161
|
+
installs the Gauge skill, which directs agents to the installed CLI's instructions.
|
|
162
|
+
|
|
163
|
+
Execution backend is independent of definition source. Resolve an explicit
|
|
164
|
+
`--local`/`--hosted` flag, then `.gauge/config.json`, user preference, and finally
|
|
165
|
+
local. Signing in does not change the backend. `gauge.json` is the shared project
|
|
166
|
+
manifest for named inputs, optional case discovery, and GitHub checks. It never
|
|
167
|
+
contains the execution preference. Git is optional for workspace initialization.
|
|
168
|
+
|
|
169
|
+
| Workflow | Local | Hosted |
|
|
170
|
+
|---|---|---|
|
|
171
|
+
| Skill, npm package (CLI or SDK), and `files` inputs | yes | yes |
|
|
172
|
+
| Docs supplied as `files` inputs | yes | yes |
|
|
173
|
+
| Eval and Agent Preference cases in `gauge-cases/` | yes (you read the sessions) | yes (judged) |
|
|
174
|
+
| `run`, `run list`, `run compare --local`, `models list`, `cleanup`, `feedback` | yes | yes |
|
|
175
|
+
| Binary, website, and MCP inputs | no | yes |
|
|
176
|
+
| Saved eval sets and preference prompts | yes | yes |
|
|
177
|
+
| Named project inputs and previews on saved measurements | yes | use case files/specs |
|
|
178
|
+
| Findings and summary queries | yes, supported fields | yes |
|
|
179
|
+
| Personas, recurring schedules, PR checks | no | yes |
|
|
180
|
+
| Optimizations (managed docs rewriting) | no | yes |
|
|
181
|
+
| Stored credentials for MCP servers and connections | no | yes |
|
|
182
|
+
| Additional reporting datasets, dashboards, org settings | no | yes |
|
|
183
|
+
|
|
184
|
+
Local runs need Docker and the agent's own login (`claude` or
|
|
185
|
+
`ANTHROPIC_API_KEY` for Claude Code; `codex login` for Codex, Pi, and
|
|
186
|
+
OpenCode). They do not need a Gauge account. Hosted runs need a Gauge account
|
|
187
|
+
(`gauge onboard`) and use credits. In local mode, a hosted-only command says so
|
|
188
|
+
and prints the next step.
|
|
189
|
+
|
|
190
|
+
For coding agents, we recommend the
|
|
191
|
+
[`gauge-agents` skill](https://github.com/gauge-sh/gauge-skills/tree/main/gauge-agents).
|
|
192
|
+
It makes Gauge CLI guidance and repository setup workflows discoverable in future
|
|
193
|
+
sessions. Install it separately from repo setup, running from your project root:
|
|
194
|
+
|
|
195
|
+
```sh
|
|
196
|
+
gauge install-skill
|
|
197
|
+
```
|
|
198
|
+
|
|
199
|
+
In a terminal, select your coding agent and installation scope in the installer.
|
|
200
|
+
Project scope adds skill files to this repo; global scope makes the skill available
|
|
201
|
+
across projects. The command runs the pinned `skills@1.5.0` installer through npx
|
|
202
|
+
and forwards its prompts and output. When a coding agent runs it, the command
|
|
203
|
+
installs for that agent at project scope without prompts and prints a short summary.
|
|
204
|
+
To choose the agent or scope explicitly:
|
|
20
205
|
|
|
21
206
|
```sh
|
|
22
|
-
|
|
207
|
+
gauge install-skill --agent claude-code --yes # this project
|
|
208
|
+
gauge install-skill --agent codex --global --yes # across projects
|
|
23
209
|
```
|
|
24
210
|
|
|
211
|
+
Repeat `--agent` to target multiple agents. `--yes` requires an explicit agent
|
|
212
|
+
target; `--copy` copies files instead of creating symlinks.
|
|
213
|
+
|
|
25
214
|
## Usage
|
|
26
215
|
|
|
27
216
|
```sh
|
|
28
217
|
gauge status # the org at a glance
|
|
29
|
-
gauge
|
|
30
|
-
gauge preference list # Agent Preference prompts
|
|
218
|
+
gauge eval list # saved eval sets
|
|
219
|
+
gauge preference list # saved Agent Preference prompts
|
|
31
220
|
gauge models list # supported model × harness × provider targets
|
|
32
|
-
gauge
|
|
33
|
-
gauge
|
|
34
|
-
gauge
|
|
221
|
+
gauge eval edit <id> --cadence weekly --samples 3 # when and how much it runs
|
|
222
|
+
gauge run --eval <id> # launch now (ORG-funded, credit-gated)
|
|
223
|
+
gauge run watch <run-id> # follow a run as the agent works
|
|
35
224
|
gauge stats rankings -o json # brand rankings (share of voice) as JSON
|
|
36
225
|
gauge query -d ecosystem -m installs,install_rate -o json # a flexible report
|
|
37
226
|
```
|
|
@@ -39,27 +228,28 @@ gauge query -d ecosystem -m installs,install_rate -o json # a flexible report
|
|
|
39
228
|
`gauge instructions` prints a Gauge overview, CLI setup and usage rules, and
|
|
40
229
|
roughly 100 words of essential guidance per module. Each summary points to
|
|
41
230
|
`gauge instructions <topic>` for the full page. `gauge --help` points agents at
|
|
42
|
-
it
|
|
43
|
-
|
|
44
|
-
|
|
45
|
-
|
|
46
|
-
Use `gauge instructions
|
|
47
|
-
model selection, observable judging criteria, and a complete
|
|
48
|
-
Use `gauge instructions preference` to
|
|
49
|
-
|
|
231
|
+
it. Install the Gauge skill with `gauge install-skill` to make this guidance
|
|
232
|
+
discoverable in future sessions. Read instructions from the installed CLI so
|
|
233
|
+
they match its commands.
|
|
234
|
+
|
|
235
|
+
Use `gauge instructions run` before writing evals or starting runs: it covers
|
|
236
|
+
developer tasks, model selection, observable judging criteria, and a complete
|
|
237
|
+
migration example. Use `gauge instructions eval` and `gauge instructions preference` to save eval sets and
|
|
238
|
+
preference prompts, and `gauge instructions session-analysis` to investigate
|
|
239
|
+
product choices, source exposure, and interventions, from individual traces to
|
|
240
|
+
larger cohorts.
|
|
50
241
|
Use `gauge instructions optimization` for the full optimization workflow.
|
|
51
242
|
`gauge instructions --list` shows available topics.
|
|
52
|
-
`gauge instructions evals --install` installs the eval guidance as a local skill.
|
|
53
243
|
|
|
54
244
|
Run `gauge --help` (or `gauge <command> --help`) for the full command set:
|
|
55
|
-
|
|
56
|
-
|
|
245
|
+
run, eval, preference, models, personas, skills, mcp, repos, optimizations,
|
|
246
|
+
brands, tags, actions, dashboards, members, org, provider keys,
|
|
57
247
|
billing, usage, stats, query, and more.
|
|
58
248
|
|
|
59
249
|
## Run committed Markdown evals
|
|
60
250
|
|
|
61
251
|
File evals keep the task, run settings, and judged criteria in Git. Create
|
|
62
|
-
`gauge-
|
|
252
|
+
`gauge-cases/healthcheck.md` with YAML frontmatter and a Markdown task body:
|
|
63
253
|
|
|
64
254
|
```markdown
|
|
65
255
|
---
|
|
@@ -76,19 +266,26 @@ config:
|
|
|
76
266
|
Add a /health endpoint and run the relevant app checks.
|
|
77
267
|
```
|
|
78
268
|
|
|
79
|
-
Check your setup, then
|
|
269
|
+
Check your setup, then preview and run the current working-tree cases and inputs:
|
|
80
270
|
|
|
81
271
|
```sh
|
|
82
|
-
gauge
|
|
83
|
-
gauge
|
|
84
|
-
gauge
|
|
85
|
-
gauge
|
|
86
|
-
gauge run
|
|
272
|
+
gauge run --verify # local; includes uncommitted working-tree edits
|
|
273
|
+
gauge run --verify -o json # versioned global/per-input yes/no/unknown checklist
|
|
274
|
+
gauge run --dry-run
|
|
275
|
+
gauge run --yes
|
|
276
|
+
gauge run wait <request-id> -o json
|
|
87
277
|
```
|
|
88
278
|
|
|
89
|
-
Cases default to `gauge-
|
|
90
|
-
that convention for `
|
|
91
|
-
|
|
279
|
+
Cases default to `gauge-cases/**/*.md`. `--files` accepts quoted paths/globs and overrides
|
|
280
|
+
that convention for `--dry-run` and `run`. Interactive CLI launches snapshot current
|
|
281
|
+
working-tree definitions in either backend, including uncommitted edits. Selected
|
|
282
|
+
filesystem inputs are packaged for local execution or hosted upload; explicit remote
|
|
283
|
+
artifacts retain their declared source. GitHub checks fetch committed definitions.
|
|
284
|
+
Content digests identify the exact bytes separately from Git provenance.
|
|
285
|
+
Preview `inputSnapshots` includes the selected source `path` and
|
|
286
|
+
`gaugeArchive: { sha256, bytes }`. Those values describe Gauge's delivery archive,
|
|
287
|
+
which contains the selected files. For npm inputs it contains the `.tgz`; the raw
|
|
288
|
+
tarball has a different hash and size. Each launch takes a fresh snapshot.
|
|
92
289
|
Each case owns its agents, models, consumer repository/ref, Persona and criteria.
|
|
93
290
|
The optional `config.inputs` selects named inputs; all attach by default.
|
|
94
291
|
|
|
@@ -107,7 +304,7 @@ account settings are outside this command's scope and do not create unknown rows
|
|
|
107
304
|
Exit 1 means at least one `no`; exit 0 means no definite static problem was found.
|
|
108
305
|
No workflow code is executed.
|
|
109
306
|
|
|
110
|
-
`gauge
|
|
307
|
+
`gauge run --verify -o json` writes exactly one JSON object to stdout, including on
|
|
111
308
|
validation failure. The versioned report provides:
|
|
112
309
|
|
|
113
310
|
- `version: 1` and `mode: "local-static"` identify the report contract and scope.
|
|
@@ -125,7 +322,7 @@ validation failure. The versioned report provides:
|
|
|
125
322
|
For example, an agent can select unresolved checks with:
|
|
126
323
|
|
|
127
324
|
```sh
|
|
128
|
-
gauge
|
|
325
|
+
gauge run --verify -o json | jq '{status, facts, global: [.global[] | select(.state != "yes")], inputs: [.inputs[] | {name, checks: [.checks[] | select(.state != "yes")]}]}'
|
|
129
326
|
```
|
|
130
327
|
|
|
131
328
|
An optional `gauge.json` declares inputs produced by your normal CI or committed
|
|
@@ -271,10 +468,13 @@ rerun Gauge to test the latest build.
|
|
|
271
468
|
With CLI 0.18.0 or later, supply built inputs directly to run without the GitHub App:
|
|
272
469
|
|
|
273
470
|
```sh
|
|
274
|
-
gauge --org my-org
|
|
275
|
-
gauge --org my-org run
|
|
471
|
+
gauge --org my-org run --input product-analytics=./dist/skill.zip --yes -o json > request.json
|
|
472
|
+
gauge --org my-org run wait "$(jq -er '.runRequestId' request.json)" --check -o json > results.json
|
|
276
473
|
```
|
|
277
474
|
|
|
475
|
+
`--check` fails the job unless the verdict is `PASSED` or `MEASURED`. Without
|
|
476
|
+
it, `run wait` exits 0 once the request settles, whatever the verdict.
|
|
477
|
+
|
|
278
478
|
Set `GAUGE_API_TOKEN` as a CI secret and replace `my-org` with your workspace slug.
|
|
279
479
|
This can run immediately after your build in the same CI job.
|
|
280
480
|
|
|
@@ -292,8 +492,9 @@ Website previews and private consumer repositories still require their configure
|
|
|
292
492
|
access. Public consumer repositories and the default workspace need no GitHub App.
|
|
293
493
|
|
|
294
494
|
`wait -o json` returns the verdict, session results, criterion explanations, and
|
|
295
|
-
session links. It exits
|
|
296
|
-
|
|
495
|
+
session links. It exits 0 once the request settles, whatever the verdict. Add
|
|
496
|
+
`--check` to exit 1 unless the verdict is `PASSED` or `MEASURED`; the results
|
|
497
|
+
JSON is still the only output. Automatic PR checks continue to use configured
|
|
297
498
|
sources; disable those checks in Gauge's GitHub settings if CI owns the trigger.
|
|
298
499
|
Save `results.json` as a CI artifact even when the check fails.
|
|
299
500
|
|
|
@@ -312,31 +513,41 @@ marker, and a readiness `pattern` with named `url` and `sha` captures. Gauge sel
|
|
|
312
513
|
candidate-specific evidence; timestamps alone cannot establish the revision.
|
|
313
514
|
Comments must belong to a unique open same-repository PR at the candidate head.
|
|
314
515
|
Previews must be public HTTPS origins, without authentication or path prefixes.
|
|
315
|
-
They are fetched live rather than snapshotted.
|
|
316
|
-
|
|
516
|
+
They are fetched live rather than snapshotted.
|
|
517
|
+
|
|
518
|
+
Canonical-site preview routing requires hosted execution. Local runs reject
|
|
519
|
+
website inputs, including `--input` overrides, before checking Docker or credentials.
|
|
520
|
+
For local content-only testing, supply docs as a `files` input or put a reachable
|
|
521
|
+
preview URL directly in the prompt. Confirm that the preview contains the edits
|
|
522
|
+
and is reachable from the agent's environment; the host's localhost is not
|
|
523
|
+
automatically reachable from a container. These alternatives change exposure and
|
|
524
|
+
do not test discovery of the canonical docs. Static verification validates
|
|
525
|
+
configuration; it does not prove backend support, delivery, or an eval pass.
|
|
317
526
|
|
|
318
527
|
Editor schemas are published at
|
|
319
528
|
`https://agents.withgauge.com/schemas/gauge/v2.7.0/gauge.json` and
|
|
320
529
|
`https://agents.withgauge.com/schemas/gauge/v2.7.0/case.json`. Associate them in editor
|
|
321
530
|
settings; `gauge.json` does not accept a `$schema` field. These release URLs are
|
|
322
|
-
immutable.
|
|
323
|
-
replacing `use` with input `type` and `path`/`paths`, and moving cases into
|
|
324
|
-
`gauge-evals/` (or setting `cases` explicitly).
|
|
531
|
+
immutable. Cases default to `gauge-cases/`, or use explicit `cases` paths.
|
|
325
532
|
|
|
326
533
|
`--org` overrides the committed organization. Root `cases` can change discovery;
|
|
327
|
-
root `version` defaults to 2. Saved evals use `gauge
|
|
328
|
-
use platform funding by default and can add `--bill-to-org`.
|
|
534
|
+
root `version` defaults to 2. Saved evals use `gauge run --eval <id>`.
|
|
329
535
|
|
|
330
536
|
## Measurements own their configuration
|
|
331
537
|
|
|
332
|
-
|
|
333
|
-
|
|
334
|
-
|
|
538
|
+
Saved eval sets and preference prompts provide a stable grouping for sessions
|
|
539
|
+
plus editable templates and defaults. Local records live in workspace SQLite;
|
|
540
|
+
hosted records live in your organization. Launches freeze identity, revision, and
|
|
541
|
+
effective settings. Renaming or editing a measurement preserves its history.
|
|
542
|
+
Local measurements run on demand; recurring schedules, personas, markets, and
|
|
543
|
+
hosted resource references require hosted execution. Select named project inputs
|
|
544
|
+
for a local measurement with repeated `--input <name>` on create or edit.
|
|
335
545
|
|
|
336
|
-
An **eval** is one prompt against one configuration
|
|
546
|
+
An **eval** is one prompt against one configuration. This hosted example uses
|
|
547
|
+
registered skills and recurring scheduling; omit those hosted features locally:
|
|
337
548
|
|
|
338
549
|
```sh
|
|
339
|
-
gauge
|
|
550
|
+
gauge eval create --name "add a healthcheck" \
|
|
340
551
|
--text "Add a /health endpoint that returns 200 OK." \
|
|
341
552
|
--criterion "Returns 200: the endpoint responds 200 with no auth" \
|
|
342
553
|
--repo https://github.com/acme/api --ref main \
|
|
@@ -345,12 +556,14 @@ gauge evals create --name "add a healthcheck" \
|
|
|
345
556
|
--cadence weekly --samples 3
|
|
346
557
|
```
|
|
347
558
|
|
|
348
|
-
`gauge
|
|
559
|
+
`gauge eval edit <id>` changes any of it later; omitted flags keep their
|
|
349
560
|
stored value, and each `--clear-*` flag empties one field. To measure the same
|
|
350
561
|
prompt against a second repository or roster, copy the eval in the app and edit
|
|
351
562
|
the copy — one eval is always one configuration.
|
|
352
563
|
|
|
353
|
-
A **preference prompt** is a reusable question.
|
|
564
|
+
A **preference prompt** is a reusable question. The following hosted example
|
|
565
|
+
uses scenarios and recurring scheduling. Locally, save a prompt and launch it
|
|
566
|
+
on demand with `gauge run --preference <id> --local`. It owns one set of run
|
|
354
567
|
settings — roster, resources, cadence, samples — and any number of
|
|
355
568
|
**scenarios**, each one repository × persona cell. Every scenario runs with
|
|
356
569
|
the prompt's settings, so one prompt measures the same question across several
|
|
@@ -374,32 +587,30 @@ A prompt with zero scenarios is a valid definition — it simply never comes due
|
|
|
374
587
|
Sessions per launch = scenarios × agent/model targets × samples. Omit the
|
|
375
588
|
run-settings flags on `create` and the org's default roster applies.
|
|
376
589
|
|
|
377
|
-
##
|
|
590
|
+
## Hosted schedules
|
|
378
591
|
|
|
379
592
|
`--cadence none` is "manual": the measurement stops coming due and runs only
|
|
380
593
|
when you launch it. There is no separate pause or enabled flag, in the CLI or
|
|
381
594
|
the app. `--cadence daily|weekly|monthly` puts it back on a rhythm, and
|
|
382
|
-
`preference get
|
|
595
|
+
`gauge preference get <id>` reads a manual prompt back as `cadence: manual`.
|
|
383
596
|
|
|
384
|
-
`gauge
|
|
597
|
+
`gauge run --eval <id>` and `gauge run --preference <id>` are the separate
|
|
385
598
|
immediate launches. They use the stored configuration and ask before spending
|
|
386
599
|
(`--yes` skips the prompt for scripts).
|
|
387
|
-
`gauge preference
|
|
388
|
-
omit it and every scenario runs.
|
|
389
|
-
|
|
390
|
-
|
|
391
|
-
credits instead. Other users always use org credits and can launch only in orgs
|
|
392
|
-
they belong to.
|
|
600
|
+
`gauge run --preference <id> --scenario <scenarioId>` (repeatable) runs a subset;
|
|
601
|
+
omit it and every scenario runs. On a recurring cadence, an eval launch or a
|
|
602
|
+
full preference launch moves the next scheduled run to one period from now. A
|
|
603
|
+
scenario subset leaves the schedule unchanged.
|
|
393
604
|
|
|
394
|
-
## Optimizations
|
|
605
|
+
## Optimizations (hosted)
|
|
395
606
|
|
|
396
607
|
Start by finding runs that encountered the surface you want to improve:
|
|
397
608
|
|
|
398
609
|
```sh
|
|
399
610
|
gauge instructions optimization
|
|
400
611
|
gauge optimizations baselines --eval ev_1 -o json
|
|
401
|
-
gauge
|
|
402
|
-
gauge
|
|
612
|
+
gauge run insight run_1 -o json
|
|
613
|
+
gauge run get run_1 --logs -o json
|
|
403
614
|
gauge optimizations start --eval ev_1 --from-run run_1 --yes
|
|
404
615
|
gauge optimizations watch cl_1
|
|
405
616
|
```
|
|
@@ -434,7 +645,7 @@ Docs `REWRITE_DOCS` proposals supply `after.pages` with URL and exact `edits`
|
|
|
434
645
|
`source`: `scrape` means current fetched content, not the captured response.
|
|
435
646
|
|
|
436
647
|
For CLI trials, inspect `cliIntervention.guidance` and `targets` in authoring
|
|
437
|
-
and use `
|
|
648
|
+
and use `gauge run get <id> --logs` for the cited command evidence. `cli-change --file change.json`
|
|
438
649
|
accepts v3 `{schemaVersion:3, target, rules, shippingIntent}` and returns a sealed
|
|
439
650
|
bundle to put in `CLI_INTERVENTION.after`. It does not save or launch. The old
|
|
440
651
|
catalog/invocation-ID workflow is retired. Skills use `skill-files` and the
|
|
@@ -455,6 +666,53 @@ spending and `trials add --idempotency-key` when retrying single-trial writes.
|
|
|
455
666
|
Adoption selects a winner without changing the baseline or deploying the change;
|
|
456
667
|
`handoff` supplies instructions to implement it in your source repository.
|
|
457
668
|
|
|
669
|
+
## Compare runs
|
|
670
|
+
|
|
671
|
+
`gauge run compare` answers before/after questions in one call. Each side takes
|
|
672
|
+
repeatable `key=value` selectors: `request=`, `batch=`, `eval=`, `revision=`,
|
|
673
|
+
`experiment=`, `trial=`, `agent=`, `model=`, `since=`, and `until=`.
|
|
674
|
+
|
|
675
|
+
```sh
|
|
676
|
+
gauge run compare --baseline request=<id> --candidate request=<id>
|
|
677
|
+
gauge run compare --baseline eval=<id> --baseline until=2026-10-01 \
|
|
678
|
+
--candidate eval=<id> --candidate since=2026-10-01
|
|
679
|
+
```
|
|
680
|
+
|
|
681
|
+
It prints the pass rate, each criterion's pass rate, and median turns, cost,
|
|
682
|
+
and duration, for all runs and per agent and model. `-o json` returns the same
|
|
683
|
+
data with deltas.
|
|
684
|
+
|
|
685
|
+
With `--local` it compares local session summaries. Use saved measurement IDs
|
|
686
|
+
with `eval=` and `revision=<number>`, or select a `batch`, `agent`, `model`, and
|
|
687
|
+
time window. Compare preference runs by batch. The comparison uses current published
|
|
688
|
+
criterion judgments and recorded turns/duration; unknown metrics and unassessable
|
|
689
|
+
criteria stay out of their denominators. Completion is distinct from a judged result.
|
|
690
|
+
Inspect findings with `gauge run insight` and query capabilities with `gauge query fields`.
|
|
691
|
+
|
|
692
|
+
```sh
|
|
693
|
+
gauge run compare --local --baseline batch=<id> --candidate batch=<id>
|
|
694
|
+
```
|
|
695
|
+
|
|
696
|
+
Each run has an `evalRevision`: a hash of its prompt, repository, persona,
|
|
697
|
+
agent context, and criteria. `gauge eval edit` changes the revision of new
|
|
698
|
+
runs. Each case has its own revision. `gauge run list` shows it, and `compare`
|
|
699
|
+
warns when one eval or case ran under more than one revision; add
|
|
700
|
+
`revision=<hash>` to pick one. `compare` also warns when a side matched no
|
|
701
|
+
runs, has runs in progress, or has runs without a verdict.
|
|
702
|
+
|
|
703
|
+
Optimization trial runs test a changed configuration, so `eval=` and other
|
|
704
|
+
selectors leave them out. Add `trial=<trialId>` to compare one trial's runs,
|
|
705
|
+
including runs it reused. `gauge run list` shows the trial key and id.
|
|
706
|
+
`compare` prints the trial behind each side's trial runs. `gauge run list` and
|
|
707
|
+
`gauge run get` show a run's `optimizationTrial`.
|
|
708
|
+
|
|
709
|
+
To try other agents or sample counts once, without editing the saved measurement, pass
|
|
710
|
+
`--agent` and `--samples` to `gauge run --eval <id>` or `--preference <id>`.
|
|
711
|
+
The stored configuration stays unchanged.
|
|
712
|
+
|
|
713
|
+
`gauge run status <request-id>` prints a launch request's progress and
|
|
714
|
+
per-session results once, without waiting.
|
|
715
|
+
|
|
458
716
|
## Models, reports, and scripting
|
|
459
717
|
|
|
460
718
|
Use `gauge models list` before pinning models. It reads the organization's live
|
|
@@ -462,10 +720,23 @@ catalog and shows every selectable logical model with its harness, provider,
|
|
|
462
720
|
and effective capabilities. Open harnesses use `--agent pi:<model>` and
|
|
463
721
|
`--agent opencode:<model>` just like the frontier harnesses.
|
|
464
722
|
|
|
465
|
-
`gauge query`
|
|
466
|
-
`
|
|
467
|
-
|
|
468
|
-
|
|
723
|
+
`gauge query` uses SQLite locally and the reporting API when hosted. Both expose
|
|
724
|
+
`findings`; local also supports run summaries. Discover capabilities with
|
|
725
|
+
`gauge query datasets` and `gauge query fields --dataset findings` using the
|
|
726
|
+
selected backend. Unsupported datasets, filters, dimensions, and metrics fail
|
|
727
|
+
with actionable errors; unavailable data never becomes a fabricated zero.
|
|
728
|
+
|
|
729
|
+
```sh
|
|
730
|
+
gauge query --local --dataset findings --dimensions kind,effect --metrics finding_count,affected_runs
|
|
731
|
+
gauge query --hosted --dataset findings --dimensions kind,effect --metrics finding_count,affected_runs
|
|
732
|
+
gauge query search "authentication failed" --eval <id> --local
|
|
733
|
+
```
|
|
734
|
+
|
|
735
|
+
Findings queries use the current published analysis, so reanalysis is not double
|
|
736
|
+
counted. `finding_count` counts findings; `affected_runs` counts distinct runs.
|
|
737
|
+
Missing analysis differs from an analyzed run with no findings. Transcript search
|
|
738
|
+
scans selected evidence files and returns run/turn references. No transcript
|
|
739
|
+
index or background indexing process is maintained.
|
|
469
740
|
|
|
470
741
|
Every read supports `-o json` for scripting. `GAUGE_API_TOKEN` overrides
|
|
471
742
|
stored credentials (useful in CI). Exit codes: `0` ok · `1` error ·
|
|
@@ -509,13 +780,60 @@ preference:
|
|
|
509
780
|
```
|
|
510
781
|
|
|
511
782
|
Write both brands in the prompt itself; metadata is only for analysis and never
|
|
512
|
-
rewrites the task. `gauge
|
|
513
|
-
and preference
|
|
514
|
-
feature; CLI 0.16.0 and earlier reject them.
|
|
783
|
+
rewrites the task. `gauge run --verify`, `--dry-run`, and `gauge run` support mixed experience
|
|
784
|
+
and preference cases.
|
|
515
785
|
|
|
516
786
|
Preference uses the existing app analyst: ranked picks for open-ended cases,
|
|
517
787
|
and a winner (or neither) for head-to-head. Results are advisory. Preference-only
|
|
518
|
-
GitHub checks finish neutral; experience criteria still gate mixed
|
|
519
|
-
`gauge run
|
|
520
|
-
input, execution, and analysis errors
|
|
788
|
+
GitHub checks finish neutral; experience criteria still gate mixed case sets.
|
|
789
|
+
`gauge run wait --check` exits successfully for `MEASURED` or `PASSED`;
|
|
790
|
+
input, execution, and analysis errors exit 1. Results describe the
|
|
521
791
|
candidate revision and do not infer a base/head delta.
|
|
792
|
+
|
|
793
|
+
## Local process lifecycle
|
|
794
|
+
|
|
795
|
+
`gauge run --local` runs in the foreground. Ctrl+C (or SIGTERM) stops the
|
|
796
|
+
queue, cancels active sessions, waits for cleanup, and exits with code 130.
|
|
797
|
+
Partial results and traces remain under `.gauge/runs`. Successful sessions
|
|
798
|
+
remove their containers and temporary credentials; the Docker image stays cached.
|
|
799
|
+
|
|
800
|
+
Use `gauge run --detach --local` to keep a batch running after the CLI exits.
|
|
801
|
+
|
|
802
|
+
Use `gauge run --local --runtime-image my-runtime:dev` to select a compatible
|
|
803
|
+
custom image, including with `--dry-run` or `--detach`. Derive from the standard
|
|
804
|
+
runtime to add Python or other system dependencies; see the
|
|
805
|
+
[Python example](https://github.com/gauge-sh/gauge-cli/tree/main/examples/runtime-python).
|
|
806
|
+
Preserve the runtime's tools, contract label, PATH, host-user permissions, and
|
|
807
|
+
entrypoint behavior. Gauge validates
|
|
808
|
+
the image before accepting a batch and pins its immutable ID for queued sessions.
|
|
809
|
+
A detached supervisor drains the queue and exits when the batch settles. The
|
|
810
|
+
launch prints its batch ID and these commands, to run from the same repository:
|
|
811
|
+
|
|
812
|
+
```sh
|
|
813
|
+
gauge run status --batch <batch-id> --local
|
|
814
|
+
gauge run wait --batch <batch-id> --local
|
|
815
|
+
gauge run cancel --batch <batch-id> --local
|
|
816
|
+
```
|
|
817
|
+
|
|
818
|
+
`gauge run wait --batch <batch-id> --local` blocks until the batch settles
|
|
819
|
+
(`completed`, `canceled`, `failed`, or `interrupted`), then prints each
|
|
820
|
+
session's state and the next step. It exits 0 once settled; `--check` exits 1
|
|
821
|
+
unless every session completed.
|
|
822
|
+
|
|
823
|
+
`gauge run status <run-id> --local` inspects a single session;
|
|
824
|
+
`gauge run cancel <run-id> --local` cancels only that run. All support `-o json`.
|
|
825
|
+
Batch cancellation is persisted before workers are contacted, including during
|
|
826
|
+
startup. A pending cancellation returns nonzero; inspect status before retrying.
|
|
827
|
+
|
|
828
|
+
After an interruption, `gauge cleanup --local` retries cleanup for recorded
|
|
829
|
+
interrupted sessions or sessions with incomplete cleanup in this repository.
|
|
830
|
+
It preserves evidence and recovered work, and makes recovered results visible
|
|
831
|
+
in `gauge run list`. It leaves live sessions, other repositories, unrelated
|
|
832
|
+
containers, and cached images alone. Recovery checks Gauge's container label
|
|
833
|
+
and workspace mount before removing a remaining container. Uncertain worker
|
|
834
|
+
ownership is reported rather than signaling a potentially reused PID.
|
|
835
|
+
|
|
836
|
+
Local workers have a 30-minute turn deadline and a 60-second supervisor
|
|
837
|
+
response deadline, so losing a supervisor cannot leave an idle worker forever.
|
|
838
|
+
Cleanup cannot discover sessions whose `.gauge` records were deleted; keep
|
|
839
|
+
those records until cleanup has succeeded.
|
|
@@ -0,0 +1,2 @@
|
|
|
1
|
+
import{runLocal}from"./chunk-CIJ4U33Y.js";import"./chunk-FCK3MQPV.js";import"./chunk-RB7Q72OV.js";import"./chunk-PSBF4TEK.js";import{randomBytes}from"crypto";import{readFile,rename,rm,writeFile}from"fs/promises";import{dirname,join}from"path";var requestPath=process.argv[2];var directory=dirname(requestPath);var requestLoaded=false;try{const request=JSON.parse(await readFile(requestPath,"utf8"));if(!request||typeof request.batchId!=="string"||typeof request.gaugeDir!=="string"||!request.definition)throw new Error("Invalid batch request");requestLoaded=true;await rm(requestPath);await runLocal(request.definition,[],request.gaugeDir,()=>{},{batchId:request.batchId,credentialSelections:request.credentialSelections,measurement:request.measurement})}catch(error){process.exitCode=1;const detail=requestLoaded?String(error):"Could not read a valid background batch request. Start a new batch.";console.error(detail);const statusPath=join(directory,"status.json");const previous=await readFile(statusPath,"utf8").then(raw=>JSON.parse(raw)).catch(()=>({}));const finishedSessions=previous.finishedSessions??0;const status={...previous,state:"failed",detail,finishedSessions,queuedSessions:0,skippedSessions:Math.max(0,(previous.expectedSessions??0)-finishedSessions),supervisorPid:process.pid};const temporary=`${statusPath}.${randomBytes(6).toString("hex")}.tmp`;await writeFile(temporary,`${JSON.stringify(status,null,2)}
|
|
2
|
+
`,{mode:384});await rename(temporary,statusPath);await writeFile(join(directory,"error.txt"),detail,{mode:384})}
|