foldrun 26.3.3 → 26.3.5

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/docs/tools.md CHANGED
@@ -61,14 +61,14 @@ headers:
61
61
  | `timeout` | seconds a call may take; without one the call runs until it finishes |
62
62
  | `rate` | calls per second from one step — `5/s`, `100/m`; a call over the limit waits, it does not fail |
63
63
  | `openapi` | a URL or file; turns one generic tool into typed ones, one per operation |
64
- | `operations` | only these operations, by `operationId` or `METHOD /path`; the generic tool is withheld. It fails closed: if the document does not load or none of them resolve, the API has no tools that run, never the generic one. Keep an allowlisted document in the workspace, so a vendor renaming an operation cannot empty it |
64
+ | `operations` | only these operations, by `operationId` or `METHOD /path`; the generic tool is withheld. It fails closed: if the document does not load or none of them resolve, the API has no tools that run, never the generic one. Keep an allowlisted document in the project, so a vendor renaming an operation cannot empty it |
65
65
 
66
66
  **The body is the manual.** Everything under a `tool.md`'s frontmatter goes
67
67
  to the model with the tool, after `description:` — how to call it, what it
68
68
  refuses, what its output means, examples of a call (up to 6,000 characters;
69
69
  a longer body is cut there and says so). A single-file tool's fenced program
70
70
  is code, not guidance, and stays out. This holds for script and HTTP tools
71
- alike, workspace or library.
71
+ alike, project or library.
72
72
 
73
73
  The model never sees a credential — the host substitutes it as the request
74
74
  goes out. The method allowlist protects against the agent; the key's own scopes
@@ -167,7 +167,7 @@ holds its value — bind a key you would not hand the agent.
167
167
 
168
168
  Resolution is **nearest wins**: an agent's own beats its project's, which beats
169
169
  the account library. So a shared integration is defined once for the whole
170
- account and rotating its credential is a single edit — and a workspace can
170
+ account and rotating its credential is a single edit — and a project can
171
171
  still shadow it by defining the same name locally.
172
172
 
173
173
  The Library page lists, for every shared file, which agents granted it and
@@ -183,7 +183,7 @@ gallery is a shelf beneath your account library, laid down on the box at
183
183
  boot, so a platform upgrade reaches every account at once.
184
184
 
185
185
  **Installing** is still there, and it means something narrower now: it
186
- copies the tool or skill into your account library (or one workspace), and
186
+ copies the tool or skill into your account library (or one project), and
187
187
  your copy shadows the platform's by name. A copy you have **not edited**
188
188
  follows the gallery: every platform deploy brings it to the current
189
189
  version. A copy you **have** edited is yours and is never changed under
@@ -199,7 +199,7 @@ not one of the skills an agent with no `skills:` list inherits.
199
199
  `foldrun run` and `foldrun eval` fetch the gallery before they start and keep
200
200
  a copy under `~/.foldrun/gallery/<host>/`, so an agent granting `web`
201
201
  runs locally exactly as it does on a deploy, with the same lookup order: the
202
- workspace's copy, then the account library's, then the shelf. Offline, the
202
+ project's copy, then the account library's, then the shelf. Offline, the
203
203
  last copy is used and the run says so. `foldrun gallery` lists the shelf and
204
204
  marks which entries your account keeps its own copy of, `foldrun gallery pull`
205
205
  refreshes the copy by hand, and `foldrun gallery upgrade <name>` is the
@@ -465,7 +465,7 @@ None of the actions hardcodes a country. `web action=search` asks the engine in
465
465
  agent's `language:` — and for its region when the tag carries one, as
466
466
  `en-AU` or `pt-BR` — `web action=browse` reports it as the page's locale, and
467
467
  `web action=fetch` sends it as `Accept-Language`. The value cascades from the agent
468
- to its workspace to the account, and is English when nobody set it; see
468
+ to its project to the account, and is English when nobody set it; see
469
469
  [`language:`](agents#language--the-language-this-agent-works-in). A
470
470
  country the tag does not carry comes from
471
471
  [`region:`](agents#region--the-country-this-agent-works-for) — `language: en`
@@ -856,7 +856,7 @@ web:
856
856
 
857
857
  A setting the vendor has no option for fails `foldrun check`, naming the
858
858
  vendors that do — the same rule as an action. `keep` stores the vendor's
859
- context or profile id in the workspace's `state/web-browse/`, so the next run
859
+ context or profile id in the project's `state/web-browse/`, so the next run
860
860
  signs in as the last one left off; it uses the vendor's own browser context
861
861
  as it stands, so `user_agent`, `device`, `locale`, `timezone` and
862
862
  `identities` beside it are refused. ZenRows' browser refuses a changed user
@@ -922,11 +922,11 @@ exactly as it always did.
922
922
  | `delta` | with `mode=aria`: `true` returns only what changed since the last snapshot of the same page in this session, or one line saying nothing did |
923
923
  | `annotate` | with `mode=screenshot`: `true` draws a box and its number on every element that can be acted on, and lists them — the numbers are `"@e4"` |
924
924
  | `visual` | with `mode=diff`: `true` compares pixels, not words — what share changed, where, and `outputs/diff.png` with the changes in red |
925
- | `init` | workspace paths of scripts run in every page before its own |
925
+ | `init` | project paths of scripts run in every page before its own |
926
926
  | `allowed_domains` | the only domains the browser may reach — pages, scripts, fetches, sockets. From the block, a call may narrow it, never widen it |
927
927
  | `boundaries` | `true` wraps the page's words in markers with a fresh nonce |
928
928
  | `webgpu` | `true` gives the page WebGPU, on a software GPU where there is none |
929
- | `extensions` | workspace paths of unpacked Chrome extensions; the browser then runs in the step, in Chromium |
929
+ | `extensions` | project paths of unpacked Chrome extensions; the browser then runs in the step, in Chromium |
930
930
  | `ignore_https_errors` | `true` opens a page whose certificate does not check out |
931
931
  | `cpu_profile` | a `.cpuprofile` path — where the page's JavaScript spent its time, for DevTools |
932
932
  | `state_key` | the NAME of a secret whose value encrypts the `state` login at rest (AES-256-GCM) |
@@ -1233,7 +1233,7 @@ page it opened the session that belongs to one site. A `cookies=` on the call
1233
1233
  is the caller's own decision and is used as given.
1234
1234
 
1235
1235
  `web: {browse: browserbase}` still means exactly what it did, so no file needs
1236
- changing. A block may sit on the agent, the workspace's `AGENTS.md` or the
1236
+ changing. A block may sit on the agent, the project's `AGENTS.md` or the
1237
1237
  account's; the nearest one wins and replaces the others whole. **A call
1238
1238
  argument beats the file** — `browser=`, `user_agent=`, `device=`, `locale=`,
1239
1239
  `timezone=` — so the block is a default. The exceptions are what the agent
@@ -1347,14 +1347,14 @@ follows it.
1347
1347
  the page cannot know, so words a site wrote to look like instructions stay
1348
1348
  labelled as the site's. It is a label, not a wall.
1349
1349
 
1350
- `init:` runs workspace scripts in every page before the page's own — a
1350
+ `init:` runs project scripts in every page before the page's own — a
1351
1351
  stub for an analytics call, a polyfill. `extensions:` loads unpacked Chrome
1352
- extensions from the workspace; the browser then runs in the step rather than
1352
+ extensions from the project; the browser then runs in the step rather than
1353
1353
  the account's pod, in Chromium (branded Chrome no longer loads them).
1354
1354
  `webgpu: true` and `ignore_https_errors: true` do what they say. `state_key:`
1355
1355
  names a secret whose value encrypts the `state=` login at rest with
1356
1356
  AES-256-GCM, so a session cookie in `state/` is not a login anyone who can
1357
- read the workspace can use; declare the secret under `secrets:`. An
1357
+ read the project can use; declare the secret under `secrets:`. An
1358
1358
  encrypted state without its key is refused rather than started signed out.
1359
1359
 
1360
1360
  **A sign-in the model never sees.** A `login` step takes a credential
@@ -1498,7 +1498,7 @@ names it, in this run or a run next month. It is the agent's own login,
1498
1498
  made with `actions` from credentials it can see — or a login a cookie secret
1499
1499
  seeded, when `state_key` encrypts it (below). Without a key `state` refuses
1500
1500
  `cookies=` and a session they seeded, and it always refuses `storage=`,
1501
- because `state/` is a workspace file every later step can read, and a vault
1501
+ because `state/` is a project file every later step can read, and a vault
1502
1502
  secret written there in the clear would stop being one. `mode=diff` keeps its
1503
1503
  baselines in the same place, under `state/browser/diff/`, for the same
1504
1504
  reason: they must outlive the run.
@@ -1646,7 +1646,7 @@ included. Reads: `Read`/`Glob`/`Grep`, `web` search, fetch, crawl, map,
1646
1646
  extract and answer, a browse that only looks (navigates, scrolls, waits,
1647
1647
  screenshots, extracts, hovers), a GET or HEAD to an API, the `search`,
1648
1648
  `history` and `desks` tools and a consult. Writes: any file written or
1649
- edited (`state/`, `memory/`, the workspace), Bash, every script tool, an
1649
+ edited (`state/`, `memory/`, the project), Bash, every script tool, an
1650
1650
  API call that is not a GET, a web action a paid provider answers, `web
1651
1651
  action=monitor` (it keeps state), a browse that clicks, fills, types,
1652
1652
  presses, selects, uploads, signs in, accepts a dialog, solves a captcha or
@@ -1872,7 +1872,7 @@ Values are whole numbers, 1 or more; to withhold a tool, leave it out of
1872
1872
  sub-agents count toward the step that delegated: their calls pass the same
1873
1873
  check.
1874
1874
  - **A refused call counts toward nothing**, and neither does one another
1875
- check refused (a path outside the workspace, a tool not granted).
1875
+ check refused (a path outside the project, a tool not granted).
1876
1876
  - **Every step has two by default: `calls: 1000` and `repeat: 25`** (since
1877
1877
  13 Nov 2026). A model that keeps calling — the same call, or anything at
1878
1878
  all — is refused, not paid for until the step's `timeout:`. Set either key
@@ -1882,7 +1882,7 @@ Values are whole numbers, 1 or more; to withhold a tool, leave it out of
1882
1882
  the model kept calling — it is stuck*. A model told "no" that often is not
1883
1883
  going to stop by itself.
1884
1884
  - **Nearest wins, per key.** The platform's defaults, then the account's
1885
- `AGENTS.md`, then the workspace's, then the agent's own, then a flow step's
1885
+ `AGENTS.md`, then the project's, then the agent's own, then a flow step's
1886
1886
  `limits: {web.search: 10}` option. A key set nearer replaces that key;
1887
1887
  the others stand.
1888
1888
  - **On the trace.** The step lists its limits (the defaults included) when it starts, each refusal
@@ -1927,7 +1927,7 @@ Only names and descriptions sit in context; the agent reads the full file when
1927
1927
  a task matches. So an agent can carry many skills for a few tokens each — which
1928
1928
  is the reason to move a procedure out of `agent.md` in the first place.
1929
1929
 
1930
- Skills are discovered in the agent's own `skills/`, the workspace's, the
1930
+ Skills are discovered in the agent's own `skills/`, the project's, the
1931
1931
  account library, and the cross-client `.agents/skills/` convention that other
1932
1932
  tools read and write. A skill written here drops into them, and one they
1933
1933
  install is visible here. The platform's [gallery](#the-gallery--platform-maintained-tools-and-skills)
@@ -1946,8 +1946,8 @@ credential** — the vault stores the refresh recipe (`token_url`, `client_id`,
1946
1946
  `client_secret`, `refresh_token`) and the host exchanges it for a live access
1947
1947
  token immediately before every use, cached until near expiry.
1948
1948
 
1949
- Resolution is nearest-wins across the workspace and account vaults, and the run
1950
- log says which store each one came from — a workspace quietly falling back to
1949
+ Resolution is nearest-wins across the project and account vaults, and the run
1950
+ log says which store each one came from — a project quietly falling back to
1951
1951
  an account credential is the kind of thing you want to see rather than infer.
1952
1952
 
1953
1953
  ### Connecting a provider
@@ -61,9 +61,9 @@ A cap over time, reached. New runs refuse until the window turns — tomorrow,
61
61
  next week, the new month — in the account's own `timezone:`. Raise it in
62
62
  Settings or in the file, or remove the line.
63
63
 
64
- ## "was denied: … is outside this workspace"
64
+ ## "was denied: … is outside this project"
65
65
 
66
- The agent reached for a path outside its workspace — `/tmp/something`, an
66
+ The agent reached for a path outside its project — `/tmp/something`, an
67
67
  absolute path, a checkout a tool made — and was refused. The message names
68
68
  the path it probably meant. It costs a turn, nothing more; if one step
69
69
  collects many, name the exact path in its prompt (`workspace/storage/x.md`).
@@ -82,9 +82,9 @@ until it is there.
82
82
 
83
83
  ## The run is parked and nobody knows
84
84
 
85
- Check the account's `notify:` — a workspace that declared its own block
85
+ Check the account's `notify:` — a project that declared its own block
86
86
  replaced the account's whole, and may have dropped `awaiting-approval`. The
87
- dashboard banner shows every parked run across workspaces regardless.
87
+ dashboard banner shows every parked run across projects regardless.
88
88
 
89
89
  ## It ran on the wrong model
90
90
 
@@ -1,6 +1,6 @@
1
- # Workspaces
1
+ # Projects
2
2
 
3
- A workspace is a folder, and the folder is the unit of everything: it is what
3
+ A project is a folder, and the folder is the unit of everything: it is what
4
4
  you edit, deploy, permission, budget and back up. An account holds several —
5
5
  one per job, in practice — and they share nothing unless it is put in the
6
6
  account's library on purpose.
@@ -31,8 +31,8 @@ Every one of those is optional except `AGENTS.md` and at least one agent.
31
31
 
32
32
  ## The account around it
33
33
 
34
- On the platform a workspace never stands alone: it lives under an account,
35
- beside the account's own `AGENTS.md` and the library every workspace there
34
+ On the platform a project never stands alone: it lives under an account,
35
+ beside the account's own `AGENTS.md` and the library every project there
36
36
  shares.
37
37
 
38
38
  ```
@@ -45,35 +45,35 @@ my-account/
45
45
  ```
46
46
 
47
47
  `foldrun init` makes exactly that on a laptop, and `foldrun new <name>` adds
48
- another workspace to it, so the folder you edit is the folder that runs. See
48
+ another project to it, so the folder you edit is the folder that runs. See
49
49
  [the CLI reference](cli#the-shape-of-a-folder) for the commands that act on the
50
50
  whole account — `check`, `deploy`, `pull`, `status`, `workspaces` — and for the
51
51
  one scope that has no API: the account's own `AGENTS.md` is not pushed or
52
52
  pulled, only edited in Settings.
53
53
 
54
- A single folder with `agents/` and `flows/` at its root is still a workspace
54
+ A single folder with `agents/` and `flows/` at its root is still a project
55
55
  and still works everywhere; `foldrun init --flat` makes one.
56
56
 
57
- ## A new workspace starts blank
57
+ ## A new project starts blank
58
58
 
59
- **Create workspace** on the dashboard, `POST /api/workspaces` with
60
- `template: true`, and `foldrun new <name>` all make a workspace with nothing in
59
+ **Create project** on the dashboard, `POST /api/workspaces` with
60
+ `template: true`, and `foldrun new <name>` all make a project with nothing in
61
61
  it to delete: an `AGENTS.md` (plus, on a laptop, the `.gitignore` that keeps
62
62
  its secret key out of git and a `CLAUDE.md` for your coding tool). No example
63
63
  agents, no flow. `foldrun new <name> --starter` and `starter: true` still make
64
64
  the example researcher, writer and `publish` flow; `foldrun init` makes it for
65
- an account's first workspace, which the getting-started guide runs.
65
+ an account's first project, which the getting-started guide runs.
66
66
 
67
- A workspace with no agents yet deploys; `check` and the deploy say so as a
68
- warning. A flow that names an agent the workspace lacks is still an error.
67
+ A project with no agents yet deploys; `check` and the deploy say so as a
68
+ warning. A flow that names an agent the project lacks is still an error.
69
69
 
70
70
  ### Importing an agent you already have
71
71
 
72
- The agents worth having in a new workspace are often ones already written in
72
+ The agents worth having in a new project are often ones already written in
73
73
  another: **Import agent** on the Agents page (and in the empty state of a blank
74
- workspace) lists every agent in the account's other workspaces you can read,
74
+ project) lists every agent in the account's other projects you can read,
75
75
  and copies the one you pick. From a terminal, `foldrun agent import
76
- <workspace>/<agent>` does it between workspaces of an account folder, or on the
76
+ <workspace>/<agent>` does it between projects of an account folder, or on the
77
77
  platform with `--to <workspace>`; the route is `POST
78
78
  /api/workspaces/<ws>/agents/import` with `{ from, agent, as? }`.
79
79
 
@@ -81,7 +81,7 @@ What is copied is what the agent **is**: `agent.md` and its own `skills/`,
81
81
  `scripts/` and `knowledge/`. What it **accumulated** stays where it was learned:
82
82
  `memory/`, `outputs/`, `state/` and session files. A name already taken is
83
83
  refused — import it under another name with `--as` / `as`, which rewrites its
84
- `name:`. The import then says what the copy uses that this workspace lacks —
84
+ `name:`. The import then says what the copy uses that this project lacks —
85
85
  tools, skills, scripts, secrets — without refusing: set the secret, import the
86
86
  tool, and it runs. The copy is independent; editing one does not change the
87
87
  other.
@@ -91,14 +91,14 @@ other.
91
91
  Agent import works inside one account. To move a desk to another account or
92
92
  installation — or to hand someone a flow — export it as a `.zip`:
93
93
 
94
- - **A workspace:** Settings → **Export and import** → **Export workspace**.
94
+ - **A workspace:** Settings → **Export and import** → **Export project**.
95
95
  - **A flow:** **Export** on its card on the Flows page. The package holds the
96
96
  flow, every flow it runs as a step, every agent those steps name (with
97
97
  `on-fail:`, `delegate:` and the colleagues those agents consult or delegate to),
98
- and the workspace tools, skills and scripts they grant — including the
98
+ and the project tools, skills and scripts they grant — including the
99
99
  script a single-file tool's `run:` names.
100
100
  - **An agent:** **Export** on its page: its folder (what agent import copies)
101
- and the workspace tools, skills and scripts it grants. Colleagues it consults
101
+ and the project tools, skills and scripts it grants. Colleagues it consults
102
102
  are named, not carried.
103
103
 
104
104
  A package is authored source only: `AGENTS.md` and the `agents/`, `flows/`,
@@ -106,13 +106,13 @@ A package is authored source only: `AGENTS.md` and the `agents/`, `flows/`,
106
106
  `memory/`, `state/`, `outputs/`, storage, runs, `.env` files or a secret's
107
107
  value. The secrets it uses travel by **name**, so the importer knows what to
108
108
  set. An agent with no `skills:` line gets every skill in scope at run time, so
109
- its package carries every workspace skill.
109
+ its package carries every project skill.
110
110
 
111
111
  Anything else in a zip — memory, state, a `.gitignore` or `CLAUDE.md` beside a
112
112
  folder you zipped yourself — is **not imported**, and the preview lists it.
113
113
 
114
114
  To import, use **Import .zip** on the Flows or Agents page or in Settings, or
115
- **Add new → Import workspace** on the home page. You pick the file and first
115
+ **Add new → Import project** on the home page. You pick the file and first
116
116
  see a preview, with nothing written yet:
117
117
 
118
118
  - what it adds;
@@ -124,8 +124,8 @@ see a preview, with nothing written yet:
124
124
  An import never deletes anything, and it is all or nothing: a file it cannot
125
125
  write (a folder already in its place, a link on the way) refuses the whole
126
126
  import before anything is written. It lands as one revision, so History can
127
- restore what it replaced. A workspace package imported under a new name makes
128
- that workspace. A `.zip` of a workspace folder made by hand (a Mac's
127
+ restore what it replaced. A project package imported under a new name makes
128
+ that project. A `.zip` of a project folder made by hand (a Mac's
129
129
  Compress included) imports the same way.
130
130
 
131
131
  From a terminal: `foldrun export [workspace] [--flow <name> | --agent <name>]`
@@ -135,11 +135,11 @@ The routes are `GET /api/workspaces/<ws>/export` and
135
135
 
136
136
  ## `AGENTS.md` — context and defaults
137
137
 
138
- The body is prose every agent in the workspace sees before its own prompt.
138
+ The body is prose every agent in the project sees before its own prompt.
139
139
  Put the things that are true of the whole desk there: the business, the
140
140
  standing rules, what the reader is like.
141
141
 
142
- The frontmatter is the workspace's defaults:
142
+ The frontmatter is the project's defaults:
143
143
 
144
144
  ```yaml
145
145
  ---
@@ -155,12 +155,12 @@ provider: # see Providers
155
155
  ---
156
156
  ```
157
157
 
158
- `owner:` names the person accountable for the workspace's agents and flows.
158
+ `owner:` names the person accountable for the project's agents and flows.
159
159
  Each agent and flow may name its own; one that does not takes the
160
- workspace's, and a workspace that does not takes the account's.
160
+ project's, and a project that does not takes the account's.
161
161
 
162
162
  `telemetry:` sends each finished run to an OpenTelemetry collector as one
163
- trace; a workspace's block replaces the account's whole. See
163
+ trace; a project's block replaces the account's whole. See
164
164
  [Runs](runs#sending-the-trace-to-your-own-tracing).
165
165
 
166
166
  The frontmatter may also group these defaults by the job they do —
@@ -170,7 +170,7 @@ same keys at the top. See [The four jobs](four-jobs) for which key goes where.
170
170
 
171
171
  `network_enforce: true` refuses a site off a step's [`network:`](network)
172
172
  list instead of only naming it in the trace. The account's `true` holds for
173
- every workspace; a workspace can add it for itself but not drop the
173
+ every project; a project can add it for itself but not drop the
174
174
  account's. See [Network](network#report-only-then-refuse).
175
175
 
176
176
  `capture:` says what runs record of each tool call: `full` (the default —
@@ -180,18 +180,18 @@ only). See [Runs](runs#what-a-run-records).
180
180
  There is an `AGENTS.md` one level up too — the account's — and the two
181
181
  compose in a way worth learning once:
182
182
 
183
- - **Prose accumulates.** The account's body is added to every workspace's,
183
+ - **Prose accumulates.** The account's body is added to every project's,
184
184
  so a rule written at the account cannot be dropped below it.
185
- - **A frontmatter key replaces whole.** A workspace that sets `notify:`
185
+ - **A frontmatter key replaces whole.** A project that sets `notify:`
186
186
  replaces the account's `notify:` entirely — it does not merge into it. To
187
187
  inherit a key, say nothing about it.
188
188
 
189
- The second rule is the one that surprises people: a workspace that declares
189
+ The second rule is the one that surprises people: a project that declares
190
190
  its own `notify:` block silently stops receiving the account's default.
191
191
 
192
192
  ### The clock, all the way down
193
193
 
194
- `timezone:` is the one default with levels below the workspace as well as
194
+ `timezone:` is the one default with levels below the project as well as
195
195
  above it, because a flow's schedule and an agent's own calendar are both real
196
196
  things to set. Nearest wins, and every level inherits the one above it unless
197
197
  it says otherwise:
@@ -214,21 +214,21 @@ to the next level with one line in the trace rather than failing the step.
214
214
  ## Pausing runs — the kill switch
215
215
 
216
216
  When something is going wrong — an agent posting what it should not, a
217
- loop spending money — pause the workspace, or the whole account, from its
217
+ loop spending money — pause the project, or the whole account, from its
218
218
  Settings → Policies → Pause runs, `foldrun workspace pause` / `foldrun account pause`, or
219
219
  the API (`POST /api/workspaces/<ws>/pause`, `POST /api/account/pause`).
220
220
 
221
221
  While paused, nothing new starts: a schedule's fire, a webhook delivery, a
222
222
  chained flow, a person's Run and a rerun are all refused, and the
223
- workspace's Triggers page logs each refused fire as *paused by a
223
+ project's Triggers page logs each refused fire as *paused by a
224
224
  person*. A refused fire is not run later. A run that was queued before
225
225
  the pause and has not begun waits until the resume. Add `--stop` (or tick
226
226
  *Also stop the runs going now*) to stop the runs already going — and the
227
227
  queued and parked ones — as well; without it, a run already under way
228
228
  finishes.
229
229
 
230
- The account's pause holds every workspace, and only the account's resume
231
- lifts it; a workspace paused on its own stays paused when the account
230
+ The account's pause holds every project, and only the account's resume
231
+ lifts it; a project paused on its own stays paused when the account
232
232
  resumes. Who paused and why is on the page, on every dashboard page while
233
233
  the account is paused, and in the audit log (`workspace.paused`,
234
234
  `account.resumed`, …). A scheduled fire inside a pause is not reported as a
@@ -238,10 +238,10 @@ deploy never lifts it.
238
238
  ## The account library
239
239
 
240
240
  `<account>/library/` holds tools, skills, knowledge and memory shared by every
241
- workspace. Resolution is nearest-wins: an agent's own file beats the
242
- workspace's, which beats the library's. So a shared integration — the tool
241
+ project. Resolution is nearest-wins: an agent's own file beats the
242
+ project's, which beats the library's. So a shared integration — the tool
243
243
  that clones your website, the skill that explains how — is defined once, and
244
- a workspace that needs a different version shadows it by defining the same
244
+ a project that needs a different version shadows it by defining the same
245
245
  name locally.
246
246
 
247
247
  From inside a run the library is read-only, reached as `account/…`
@@ -272,17 +272,17 @@ memory an agent wrote that your deploy does not mention. Those belong to the
272
272
  platform, not to your source, and a deploy that reverted what an agent learned would
273
273
  make every run a little dumber.
274
274
 
275
- The agent engine's own files are not workspace content either. Each step
275
+ The agent engine's own files are not project content either. Each step
276
276
  runs in a sandbox on the Claude Agent SDK — the engine, whichever model
277
277
  answers — and the engine writes its settings and session transcripts in the
278
278
  agent's folder there (`agents/<name>/.claude.json`, `agents/<name>/.claude/`).
279
279
  They stay in the sandbox: never copied back, never copied in. (Your own
280
- `.claude/agents/*.md` at the workspace root is different — a Claude Code
280
+ `.claude/agents/*.md` at the project root is different — a Claude Code
281
281
  subagent you wrote, deployed as an agent.)
282
282
 
283
283
  A deploy while a run is in flight is refused, saying which runs hold it; a
284
284
  run parked at an approval gate counts as in flight. See [Deploying](deploying).
285
285
 
286
- Trying a change to a live desk without touching it is a second workspace —
286
+ Trying a change to a live desk without touching it is a second project —
287
287
  `foldrun deploy --to <workspace>-try`. Branch previews come back with the
288
288
  GitHub and GitLab app.
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "foldrun",
3
- "version": "26.3.3",
3
+ "version": "26.3.5",
4
4
  "description": "Agents are just folders. Write them, check them, run them — on your machine.",
5
5
  "license": "FSL-1.1-ALv2",
6
6
  "type": "module",
@@ -11,7 +11,7 @@
11
11
  "node": ">=22"
12
12
  },
13
13
  "dependencies": {
14
- "@foldrun/core": "^24.2.1"
14
+ "@foldrun/core": "^24.2.2"
15
15
  },
16
16
  "files": [
17
17
  "bin",