You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+4-4Lines changed: 4 additions & 4 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -31,7 +31,7 @@ bun add --dev skillgym
31
31
Create `skillgym.config.ts` in your project root, or in a parent directory that the suite can discover upward:
32
32
33
33
```ts
34
-
importtype { SkillGymConfig } from"skillgym";
34
+
importtype { SkillgymConfig } from"skillgym";
35
35
36
36
const config = {
37
37
run: {
@@ -67,7 +67,7 @@ const config = {
67
67
},
68
68
},
69
69
},
70
-
} satisfiesSkillGymConfig;
70
+
} satisfiesSkillgymConfig;
71
71
72
72
exportdefaultconfig;
73
73
```
@@ -179,13 +179,13 @@ The execution unit is one case x runner pair. `skillgym` expands the suite into
179
179
180
180
`serial` is the default. `parallel` maximizes overlap across the full matrix up to the configured cap. `isolated-by-runner` is a middle ground when you want each runner to stay ordered internally but still allow different runners to overlap.
181
181
182
-
For concurrent schedules, `run.maxParallel` defaults to `os.availableParallelism()`. This limits how many SkillGym executions are active at once; it does not pin or limit CPU cores used by an individual agent process.
182
+
For concurrent schedules, `run.maxParallel` defaults to `os.availableParallelism()`. This limits how many Skillgym executions are active at once; it does not pin or limit CPU cores used by an individual agent process.
183
183
184
184
Concurrent schedules do not copy or isolate the workspace by themselves. Overlapping executions may still interact through the same filesystem state and live runner output unless you use isolated workspaces. OpenCode, Codex, and Claude Code runtime state are isolated per execution under each artifact directory.
185
185
186
186
`run.repeat` is useful when you want stability sampling instead of a single lucky pass. Each case x runner execution keeps running until it records the requested number of successful classified results, or stops early when one repetition still fails after exhausting `run.repeatFailure` retries.
187
187
188
-
`run.repeatFailure` retries only the current repetition when it still counts as failed after result classification. SkillGym keeps all repetition and retry artifacts, averages visible duration and token metrics across the final results of completed successful repetitions, and preserves the full nested detail in `results.json`.
188
+
`run.repeatFailure` retries only the current repetition when it still counts as failed after result classification. Skillgym keeps all repetition and retry artifacts, averages visible duration and token metrics across the final results of completed successful repetitions, and preserves the full nested detail in `results.json`.
189
189
190
190
Artifacts for repeated executions are grouped under the stable case x runner directory using `repeat-N` directories, with retry sessions nested as `session-N` inside the repetition that needed recovery.
Copy file name to clipboardExpand all lines: docs/workspaces.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -8,7 +8,7 @@
8
8
-`shared`: create one workspace per suite run and reuse it across executions
9
9
-`isolated`: create a fresh workspace per case x runner execution
10
10
11
-
Use `none` when the agent should run in an existing directory. Use `shared` or `isolated` when the suite needs SkillGym to provision a workspace first.
11
+
Use `none` when the agent should run in an existing directory. Use `shared` or `isolated` when the suite needs Skillgym to provision a workspace first.
12
12
13
13
## Suite-level workspace config
14
14
@@ -61,7 +61,7 @@ Behavior:
61
61
- if omitted, none mode falls back to config `run.cwd`, then `process.cwd()`
62
62
-`templateDir` and `bootstrap` are not allowed
63
63
64
-
If a suite file does not export `workspace`, SkillGym falls back to config `run.workspace`, then to implicit `none` mode.
64
+
If a suite file does not export `workspace`, Skillgym falls back to config `run.workspace`, then to implicit `none` mode.
0 commit comments