fimo 2.10.3 → 2.10.4-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/hosting.md +1 -1
- package/assets/skills/fimo-cli/references/knowledge.md +10 -3
- package/assets/skills/fimo-studio/references/ai-chat/agents.md +1 -1
- package/assets/skills/fimo-studio/references/ai-chat/default-agents.md +5 -5
- package/dist/cli/bundle.json +1 -1
- package/dist/cli/index.js +11 -11
- package/package.json +8 -8
- package/release.json +2 -2
- package/templates/astro/package.json +3 -3
- package/templates/nextjs/package.json +4 -4
- package/templates/react-router/package.json +5 -5
- package/templates/sveltekit/package.json +3 -3
- package/templates/vite/package.json +5 -5
|
@@ -114,7 +114,7 @@ header on a static asset.
|
|
|
114
114
|
serves `/about` and `/about/` as separate URLs, which search engines index as
|
|
115
115
|
duplicates.
|
|
116
116
|
- **An empty `hosting: {}` means no platform routing**, which is different from
|
|
117
|
-
omitting `hosting` entirely — that leaves legacy SPA
|
|
117
|
+
omitting `hosting` entirely — that leaves legacy SPA behavior untouched.
|
|
118
118
|
- **Not supported:** cron jobs (use `fimo schedules`), image optimization (use
|
|
119
119
|
Fimo media), per-function runtime settings, deployment regions.
|
|
120
120
|
|
|
@@ -32,16 +32,23 @@ permission to learn and persist. A request to change site copy is not a request
|
|
|
32
32
|
Read the current knowledge, edit the complete Markdown while preserving unrelated guidance, and save:
|
|
33
33
|
|
|
34
34
|
```bash
|
|
35
|
-
fimo knowledge set --file knowledge.md --json
|
|
36
|
-
cat knowledge.md | fimo knowledge set --stdin --json
|
|
35
|
+
fimo knowledge set --file knowledge.md --expected-updated-at '<updatedAt from get>' --json
|
|
36
|
+
cat knowledge.md | fimo knowledge set --stdin --expected-updated-at '<updatedAt from get>' --json
|
|
37
37
|
```
|
|
38
38
|
|
|
39
39
|
`--file -` also reads stdin. Content is limited to 10,000 characters. An empty file explicitly clears
|
|
40
40
|
knowledge. Each save replaces the current Markdown.
|
|
41
41
|
|
|
42
|
+
Use `none` for a read whose `updatedAt` is null. On `KNOWLEDGE_CONFLICT`, read again and merge
|
|
43
|
+
your changes while preserving the newer sections; save with the revision from that read.
|
|
44
|
+
Never fetch a newer timestamp just to submit the old content or retry without the condition.
|
|
45
|
+
Fimo's structured `fimo` tool tracks the revision itself: use `knowledge get` successfully
|
|
46
|
+
before `knowledge set` in the same run and omit the flag from tool arguments.
|
|
47
|
+
|
|
42
48
|
CLI `--branch` does not isolate knowledge. Saved updates apply on the next agent context refresh.
|
|
43
49
|
For MCP, use `knowledge_get` or `knowledge_update`
|
|
44
|
-
with `projectId`; update accepts `content
|
|
50
|
+
with `projectId`; update accepts `content` and optional `expectedUpdatedAt`, the timestamp read
|
|
51
|
+
or null for an absent document. Standalone CLI and MCP calls that omit it still replace unconditionally.
|
|
45
52
|
|
|
46
53
|
Settings also offers **Draft from website**. It runs analysis in the background and returns an editable
|
|
47
54
|
draft for review. Only an explicit **Save** updates shared knowledge; drafting does not grant permission
|
|
@@ -43,4 +43,4 @@ The **+** menu lists them under **Agents**: **SEO**, **GEO**, **Content**, **Qua
|
|
|
43
43
|
|
|
44
44
|
A change on main wakes the routines it can matter to, before a publish: CMS edits, code changes, images, languages and forms; the first publish and a new domain also wake Technical SEO and Cookie consent. A small model reads each change first, so a change that matters to no routine starts nothing; schedules run anyway. A run that a change started checks only the pages it touches.
|
|
45
45
|
|
|
46
|
-
**Alone:** a fix with one right answer from the page's own words, without a message. **Asked as a decision:** a
|
|
46
|
+
**Alone:** a fix with one right answer from the page's own words, without a message, including shorter text from the same words and a rewrite that only applies the team's own rules. **Asked as a decision:** new words or a new meaning in what visitors read, which page wins a search, whether a page exists, a redirect with more than one possible target, a price or a legal page. They never remove a page or change the robots policy alone. A change a person asks for in chat is made. To stop a kind of question, tell the agent in its chat ("do this on your own from now on", "don't ask me this again"): it saves the rule in its memory and no routine of that agent asks it again. A rule never covers the robots policy, a price, a legal page or removing a page. A check that could not finish (a timeout, a 5xx, a page past the crawl limit) is retried once, then reported as incomplete, never acted on.
|
|
@@ -10,12 +10,12 @@ Plans, setup and the rules every default agent follows are in `ai-chat/agents`.
|
|
|
10
10
|
- **Image SEO:** alt text, file names, and images over 200 KB. An image in the media library is rendered through Fimo's resizing; a file of the site is uploaded first. An image on another host gets a proposal for a smaller export.
|
|
11
11
|
- **Hreflang:** on a site in more than one language, each version lists the others.
|
|
12
12
|
- **Internal linking:** orphan pages, pages more than three clicks deep, and articles with no link to a hub or product page or more than eight links. Runs on changes and every Monday.
|
|
13
|
-
- **Keyword optimization** (every Tuesday, needs Search Console): for pages ranking 8th to 20th it
|
|
13
|
+
- **Keyword optimization** (every Tuesday, needs Search Console): for pages ranking 8th to 20th it writes a title and headings from the searches that bring visitors, keeps the old words, and checks 8 weeks later whether they helped; on a clear loss it puts the old words back.
|
|
14
14
|
- **Keyword research** (2nd of each month): writes at most five briefs to the site knowledge and asks which to write; the Content agent's Writing routine writes the approved ones.
|
|
15
15
|
|
|
16
16
|
The Monday runs of Technical SEO, Meta tags and Internal linking read every page of main's hosted preview, for what no single change shows (pages out of the sitemap, shared titles, title shape, orphan and deep pages). A run that a change started checks only the pages it touches, its first run too.
|
|
17
17
|
|
|
18
|
-
Alone, it
|
|
18
|
+
Alone, it points a link at a new address only after it checked that the address shows the same content in the same language, and otherwise removes a dead link and keeps its words; it adds a link on words the text already uses to name another page, keeps the 8 most useful links on a page with too many, shortens a title or description Google cuts from the same words (in a field only search reads, added to the type when needed, when the title is also the page's heading), sets the public address when the canonicals and sitemap agree on it, and adds a 301 from a page's old address once the same page is live at the new one. A title, description, heading or link it changed is not changed again for 8 weeks (Google needs 8 to 12 weeks to show the effect), unless the change broke something, and words a person rewrote after an agent changed them stay. Keyword research never changes the site alone.
|
|
19
19
|
|
|
20
20
|
## GEO (4 routines)
|
|
21
21
|
|
|
@@ -24,7 +24,7 @@ Alone, it fixes a link only after it checked that the new address shows the same
|
|
|
24
24
|
- **AI citation optimization:** a direct answer under each content page's H1, and question headings on long pages.
|
|
25
25
|
- **AI visibility** (once a week): how often ChatGPT, Gemini, Perplexity and Google AI Mode name and cite the site for about 20 customer questions. It changes nothing.
|
|
26
26
|
|
|
27
|
-
Alone, it writes a missing llms.txt and keeps its
|
|
27
|
+
Alone, it writes a missing llms.txt and keeps its lines in step with the pages and their titles and descriptions, adds Organization, WebSite, Article and BreadcrumbList from facts the site already shows (never an invented logo, address, author or rating), and fixes landmarks and heading levels without changing a word. It asks before an opening answer, new headings, FAQ or Product data, or a change to how pages are rendered (unless one setting serves the same pages as HTML and nothing else changes). It never changes who robots.txt lets crawl, AI crawlers included, and never invents a claim; asked, it names the crawlers robots.txt keeps out.
|
|
28
28
|
|
|
29
29
|
## Content (6 routines)
|
|
30
30
|
|
|
@@ -44,7 +44,7 @@ It also measures accessibility in each page, and opens one or two pages of each
|
|
|
44
44
|
- **Mobile optimization:** a missing viewport or one that blocks zoom, sideways scrolling, images, tables and forms wider than the screen, tap targets under 24 × 24 px, text mostly under 12 px.
|
|
45
45
|
- **Page speed** (3rd of each month): Lighthouse on a phone for the home page and one page of each template.
|
|
46
46
|
|
|
47
|
-
Alone, it names a control or labels a field from words the page shows, fixes heading levels, tabindex and a removed focus outline, lets visitors zoom again,
|
|
47
|
+
Alone, it names a control or labels a field from words the page shows, fixes heading levels, tabindex and a removed focus outline, lets visitors zoom again, keeps media inside the screen by its own width, gives small tap targets 24 × 24 px, and moves a failing color to a passing shade the site's palette already has. Any other color, size or layout change is asked as a decision. Image alt text stays with SEO.
|
|
48
48
|
|
|
49
49
|
## Compliance (3 routines)
|
|
50
50
|
|
|
@@ -54,4 +54,4 @@ It can open a sample of pages with no consent given and record the cookies, thir
|
|
|
54
54
|
- **Cookie consent:** trackers and third-party cookies before a choice, a missing banner, a banner with no way to refuse.
|
|
55
55
|
- **Privacy and legal pages:** a privacy page exists, names the services loaded and the kinds of field the forms collect, and was updated in the last two years.
|
|
56
56
|
|
|
57
|
-
Alone, it
|
|
57
|
+
Alone, it applies the brand rules where the meaning stays: a term they replace, a claim they ban that the sentence does not need, their voice for a sentence that keeps every name and number, and spelling. It proposes a sentence whose meaning would change, the consent setup and every legal page, with placeholders for legal details, and never states that a site is compliant. It does not ask the team to source its own people's quotes or its own figures.
|