fimo 2.3.0-staging.1 → 2.3.0-staging.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/assets/skills/fimo-cli/references/agents.md +9 -8
- package/dist/cli/bundle.json +1 -1
- package/dist/cli/index.js +12 -12
- package/package.json +8 -8
- package/templates/astro/package.json +2 -2
- package/templates/nextjs/package.json +3 -3
- package/templates/react-router/package.json +4 -4
- package/templates/sveltekit/package.json +2 -2
- package/templates/vite/package.json +4 -4
|
@@ -38,7 +38,7 @@ fimo agents secrets remove <name> <KEY> [-C <dir>]
|
|
|
38
38
|
```
|
|
39
39
|
|
|
40
40
|
- **`list-templates`** — list the bundled starter templates with the first line of each template's `GOAL.md` as a summary. Ships with the CLI binary; no API call. The five bundled templates are `_default`, `seo-audit`, `content-translator`, `broken-link-sweeper`, `post-deploy-check`.
|
|
41
|
-
- **`create <name>`** — scaffold `.fimo/agents/<name>/` from a template (default `_default`, a minimal valid bundle with a manual trigger). `--description <text>` writes a one-line tagline into the scaffolded `config.yaml` (surfaces in the dashboard's agents list). When run **without** `-C`, it requires the cwd to be inside a real Fimo project (walks up for `.fimo/project.json` or legacy `.fimo.settings.json`) and errors out cleanly otherwise, so it never leaks an agent bundle into a stray cwd. If the template declares capabilities, an install-consent flow runs after the copy (`--quiet` skips it; critical caps still abort).
|
|
41
|
+
- **`create <name>`** — scaffold `.fimo/agents/<name>/` from a template (default `_default`, a minimal valid bundle with a manual trigger). `--description <text>` writes a one-line tagline into the scaffolded `config.yaml` (surfaces in the dashboard's agents list). When run **without** `-C`, it requires the cwd to be inside a real Fimo project (walks up for `.fimo/project.json` or legacy `.fimo.settings.json`) and errors out cleanly otherwise, so it never leaks an agent bundle into a stray cwd. If the template declares capabilities, an install-consent flow runs after the copy (`--quiet` skips it; critical caps still abort). Keep the scaffolded `runtime` block as is unless the author asks for a different credit limit; see the `runtime` reference below before adding any limit key.
|
|
42
42
|
- Every generated `config.yaml` includes `model: openai/gpt-5.6-luna`. Change that top-level value to another supported Fimo text model when needed; bundles without `model` keep the Luna fallback. Deploy after changing it because cloud runs use the deployed bundle.
|
|
43
43
|
- **`list`** — list the agents declared under `<project>/.fimo/agents/` (or legacy root `agents/`), with each one's trigger types and capability count. Local — the bundle list is the source of truth even if the project isn't linked yet.
|
|
44
44
|
- **`validate [name]`** — run the full validation pipeline: parse → capability vocabulary → composition (cycle / missing sub-agents / invalid grants) → semantic + Fimo project config policy checks. Omit `name` to validate every bundle. `--json` emits a structured `ValidationReport` for CI; exit code is non-zero on any failure.
|
|
@@ -64,19 +64,20 @@ runtime:
|
|
|
64
64
|
dev: false
|
|
65
65
|
long_running: false
|
|
66
66
|
max_credits: 5000
|
|
67
|
-
max_steps: 40
|
|
68
67
|
```
|
|
69
68
|
|
|
70
69
|
- `cloud` / `dev` — declared boolean environment flags for where the bundle may run.
|
|
71
70
|
- `long_running` — same capability as the `runtime:long_running` token: lets the run exceed the default time limit.
|
|
72
|
-
- `max_credits` — the run's own credit ceiling (integer, credits ≥ 1). The default scaffold
|
|
73
|
-
`max_credits: 5000` as a visible starter limit
|
|
74
|
-
credits at admission.
|
|
75
|
-
- `max_steps` — cap on model-loop steps for one run (integer ≥ 1).
|
|
76
|
-
|
|
71
|
+
- `max_credits` — the run's own credit ceiling (integer, credits ≥ 1) and the one limit to tune. The default scaffold
|
|
72
|
+
template ships `max_credits: 5000` as a visible starter limit; raise, lower, or remove it to fit the agent's work.
|
|
73
|
+
Unset: the run may spend up to the organization's available credits at admission.
|
|
74
|
+
- `max_steps` — advanced cost-management knob, extremely rare: a cap on model-loop steps for one run (integer ≥ 1).
|
|
75
|
+
Leave it unset unless the author explicitly wants a step-based bound on top of credits. Do not add it when
|
|
76
|
+
scaffolding a bundle, and do not reach for it to control spend — tune `max_credits` instead. Unset: no step cap;
|
|
77
|
+
`max_credits` (or the organization's balance) and the plan's agent-hours bound the run.
|
|
77
78
|
|
|
78
79
|
All five keys are optional; an absent `runtime` block behaves as if every key were unset. There is no platform step cap
|
|
79
|
-
and no
|
|
80
|
+
and no platform per-run cost ceiling — the only stops are these author-owned limits, the organization running out of
|
|
80
81
|
credits (`insufficient_credits`), and the agent-hours allowance. A limit that fires records a truthful terminal reason
|
|
81
82
|
(`max_steps`, `budget_exceeded`, `insufficient_credits`, or `finish_run_missing` when a run ends without calling
|
|
82
83
|
`finish_run`), readable via `fimo agents report` and the Studio run page.
|