## Why
The build derived `<title>` from the **sidebar label** and the
description from the **first paragraph clipped to 220 characters**. Both
are the wrong source. A sidebar label is written for a 240px-wide nav
that supplies its own context; a search result has none.
On production that produced:
- **7 pages titled "Overview"**, and 32 pages sharing a title with
another page
- **109 of 193 descriptions cut off mid-word**, and 174 over 160
characters (where Google truncates)
```
/guides/welcome/key-concepts/
BEFORE (220): ...and if you misunderstand them early, the product wi
```
Near-identical titles are the weak-differentiation signal behind
"crawled — currently not indexed", which is the exact symptom on this
property.
## What changed
`seo_title` and `seo_description` frontmatter on **every one of the 192
pages**, written by hand against each page's real content and optimised
for the click — lead with what the reader can do, name the concrete
things covered, no keyword stuffing, no repeating the title.
| | Before | After |
|---|---|---|
| Duplicate titles | 32 pages | **0** |
| Duplicate descriptions | — | **0** |
| Descriptions cut mid-word | 109 | **0** |
| Descriptions over 160 chars | 174 | **0** |
| Title length (incl. suffix) | up to "Overview" | **26–60** |
| Description length | 31–240 | **122–158** |
**No prose was rewritten.** I measured content overlap between the
colliding pairs first — only 17–23%, so these were always distinct pages
wearing the same label, not duplicate content. The `docs/` diff is **564
insertions, zero deletions**.
### Examples
```
BEFORE — all seven identical: "Overview | Paperclip Docs"
AFTER: "Adapters Overview" "API Overview" "CLI Overview"
"Deployment Overview" "Bundled Skills" "Optional Skills"
"Experimental Features"
BEFORE — both identical: "Issues | Paperclip Docs"
AFTER: "Issues: Creating and Tracking Work" /guides/day-to-day/issues/
"Issues API" /reference/api/issues/
```
```
/guides/getting-started/five-minute-path/
BEFORE (220): The shortest route from a running Paperclip instance to an agent that has
just completed work for you. After Paperclip is installed, the path takes
about 5 minutes: create a company, hire a CEO agent, approve its first s
AFTER (131): Install Paperclip, create a company, hire a CEO agent, and approve its
first strategy — a working AI company in about five minutes.
```
## Plumbing
- **`site/build-release.mjs`** prefers the authored fields; the derived
path stays only as a fallback the CI gate prevents ever reaching.
`<title>`, `<meta name="description">`, and the `TechArticle` JSON-LD
all read from the same source.
- **`site/app.js`** uses the same fields, so a client-side transition
cannot leave a different title on screen than the crawler was served.
Verified in-browser: navigating `/guides/day-to-day/issues/` →
`/reference/api/issues/` swaps to exactly the server-rendered values, no
JS errors.
- **`scripts/verify-seo-metadata.mjs`** (new) enforces presence,
uniqueness, length, complete-sentence endings, and the
frontmatter-safety rules. Wired into `docs:test` and CI as its own step.
I negative-tested the gate rather than assuming it works — it correctly
fails on a duplicate title, a mid-word description, and a missing field,
and passes again once restored.
## Automating it going forward
`skills/sync-docs/SKILL.md` now:
- gains an **"Authored SEO metadata"** section: why the fields exist,
the enforced rules as a table, and how to actually write them (title
names the subject not the nav position; description writes for the
click; never generate mechanically from the first paragraph — *that is
the fallback the gate exists to prevent*).
- requires both fields in the **page-authoring instruction** given to
every subagent, with guidance on when to update an existing description
versus leave it.
- runs `docs:test:seo-metadata` and `docs:test:crawlable-links` as
**fail-loud gates in Phase 7** before commit.
- documents all six build gates in the files/gates section.
So once this lands, a new page cannot ship without hand-written
metadata, and a title collision cannot reach production.
## Verified
All six verifiers pass:
```
Static route verification passed
Asset fingerprint verification passed
Hierarchical Skills nav verification passed
Skill source block verification passed (11 skill pages, 1 control page)
Crawlable-link contract passed: 40886 internal links across 194 pages resolve in one hop
Authored-SEO contract passed: 192 pages carry a unique, hand-written seo_title and seo_description
```
🤖 Generated with [Claude Code](https://claude.com/claude-code)
---------
Co-authored-by: Claude Opus 5 (1M context) <[email protected]>
5.5 KiB
paperclip_version, seo_title, seo_description
| paperclip_version | seo_title | seo_description |
|---|---|---|
| v2026.618.0 | Pi Adapter | Run the Pi coding agent CLI on the Paperclip host, with its built-in tool set — read, bash, edit, write, grep, find, ls — and provider configuration. |
Pi
pi_local runs the Pi coding agent CLI on the same machine as Paperclip. Use it when you want Pi's built-in tool set (read, bash, edit, write, grep, find, ls), provider/model routing, and session resume across heartbeats.
When To Use
- You already use the Pi CLI locally.
- You want provider/model routing in Pi's
--provider <name> --model <id>format. - You want Paperclip to resume Pi sessions across heartbeats via
--session. - You need Pi's tool set available to the agent.
When Not To Use
- The runtime lives behind a webhook or API. Use OpenClaw Gateway or HTTP.
- You only need a one-shot shell command. Use Process.
- Pi CLI is not installed on the machine.
Common Fields
| Field | Required | Notes |
|---|---|---|
cwd |
no | Absolute working directory. Recommended. Created when permissions allow; otherwise falls back to the process working directory. |
model |
yes | Pi model id in provider/model format (e.g. xai/grok-4). |
thinking |
no | Thinking level passed to Pi. Accepts off, minimal, low, medium, high, xhigh. |
promptTemplate |
no | User prompt template passed via the -p flag. |
instructionsFilePath |
no | Absolute path to a Markdown instructions file appended to Pi's system prompt via --append-system-prompt. |
command |
no | Defaults to pi. Override only for a non-default executable path. |
env |
no | Environment variables. Secret refs supported. |
timeoutSec |
no | Run timeout in seconds. 0 means no timeout. |
graceSec |
no | SIGTERM grace period before a forced stop. |
Note: Paperclip requires an explicit
modelvalue forpi_localagents. Usepi --list-modelsto list valid options on your machine.
Session Persistence
Pi sessions are stored under ~/.pi/paperclips/ and resumed with --session on the next heartbeat. The session is scoped per agent so multiple Paperclip agents can share a host without stepping on each other.
Execution Details
- Agent instructions are appended to Pi's system prompt via
--append-system-prompt. - The user task prompt is sent via
-p. - All tools (
read,bash,edit,write,grep,find,ls) are enabled by default.
Custom Providers / Gateway Routing
Pi has no base-URL flag or env var, so the only way to point it at a custom or remote OpenAI- or Anthropic-compatible endpoint is a models.json file in its agent-config directory. Paperclip wires this up for you through the PAPERCLIP_PI_PROVIDERS environment variable.
Set PAPERCLIP_PI_PROVIDERS to a JSON object in Pi's models.json providers shape — keyed by provider name, with each entry describing how to reach the endpoint and which models it serves. Paperclip reads it server-side, writes {"providers": ...} to a managed models.json in a temporary agent-config dir, and points PI_CODING_AGENT_DIR at that dir for the run. (For remote execution targets, the dir is shipped to the sandbox and PI_CODING_AGENT_DIR is repointed at the in-sandbox copy.)
Because Pi resolves --provider <name> --model <id> by an exact (provider, model id) match against that registry, each provider must enumerate its models explicitly.
{
"tensorix": {
"baseUrl": "http://gateway.example.svc.cluster.local:8080/anthropic",
"apiKey": "{env:ANTHROPIC_API_KEY}",
"api": "anthropic-messages",
"models": [
{
"id": "deepseek/deepseek-chat-v3.1",
"name": "DeepSeek v3.1",
"reasoning": false,
"input": ["text"],
"cost": { "input": 0, "output": 0, "cacheRead": 0, "cacheWrite": 0 },
"contextWindow": 128000,
"maxTokens": 32000
}
]
}
}
Notes on behavior:
- Secrets stay declarative. An
apiKeymay be a literal value or an{env:VAR}placeholder. Placeholders are expanded server-side against the runenvand the Paperclip process env, so gateway keys are baked intomodels.jsonwhere the value is reliably present. An unresolvable placeholder (unset, or set to an empty string) is left intact for Pi to try itself. - The managed dir replaces the host agent dir. Credentials travel inside the providers config (via the literal or expanded
apiKey), not from the host's~/.pi/agent. - Misconfiguration is surfaced, not silent. Invalid JSON, a non-object value, or provider entries that are not themselves objects are ignored with a diagnostic note rather than failing later with an opaque model-not-found error. Leaving the variable unset (or empty) is the normal "feature off" case.
Pi then routes --provider <name> --model <id> by matching the provider key (e.g. tensorix) and a model id you listed (e.g. deepseek/deepseek-chat-v3.1) against this registry.
Example
{
"adapterType": "pi_local",
"adapterConfig": {
"cwd": "/Users/me/projects/paperclip-workspace",
"model": "xai/grok-4",
"thinking": "high",
"promptTemplate": "You are the engineering lead. Work carefully and report progress.",
"env": {
"XAI_API_KEY": {
"type": "secret_ref",
"secretId": "xai-api-key",
"version": "latest"
}
},
"timeoutSec": 300,
"graceSec": 15
}
}