## 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]>
5.6 KiB
paperclip_version, seo_title, seo_description
| paperclip_version | seo_title | seo_description |
|---|---|---|
| v2026.626.0 | Hermes Adapter | Run Hermes Agent by Nous Research on the Paperclip host, with persistent memory and a 30-plus tool suite available to the agent on every run. |
Hermes
hermes_local runs Hermes Agent — a full-featured AI agent by Nous Research — on the same machine as Paperclip. Use it when you want persistent memory, a 30+ tool suite, 80+ loadable skills, multi-provider model routing, and MCP client support in a single adapter.
When To Use
- Hermes is a built-in adapter — no plugin install required. Paperclip ships
hermes_localout of the box. - You need persistent memory, FTS5 session search, or sub-agent delegation.
- You want to route to multiple inference providers (Anthropic, OpenRouter, OpenAI, Nous, OpenAI Codex, Copilot, Copilot ACP, HuggingFace, ZAI, Kimi Coding, MiniMax, Kilocode).
- You want filesystem checkpoints for rollback safety.
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.
- Hermes Agent is not installed on the machine (requires Python 3.10+).
Common Fields
Core
| Field | Required | Default | Notes |
|---|---|---|---|
provider |
no | auto |
API provider. Accepts auto, openrouter, nous, openai-codex, copilot, copilot-acp, anthropic, huggingface, zai, kimi-coding, minimax, minimax-cn, kilocode. Usually leave this on auto — Hermes infers it from your model or ~/.hermes/config.yaml. |
timeoutSec |
no | 1800 |
Execution timeout in seconds. |
graceSec |
no | 10 |
Seconds to wait after SIGTERM before killing the Hermes process. |
Tools
| Field | Required | Default | Notes |
|---|---|---|---|
toolsets |
no | all | Comma-separated list of toolsets, such as terminal,file,web. |
maxTurnsPerRun |
no | — | Optional Hermes --max-turns limit for tool-calling iterations. |
Session & workspace
| Field | Required | Default | Notes |
|---|---|---|---|
persistSession |
no | true |
Resume sessions across heartbeats via Hermes --resume. |
worktreeMode |
no | false |
Git worktree isolation. |
checkpoints |
no | false |
Enables filesystem checkpoints for rollback. |
Advanced
| Field | Required | Default | Notes |
|---|---|---|---|
verbose |
no | false |
Verbose output. |
quiet |
no | true |
Clean output — no banner or spinner. |
promptTemplate |
no | built-in | Custom prompt template (see below). |
paperclipApiUrl |
no | — | Optional API base override. Defaults to PAPERCLIP_API_URL. |
Model Selection
Hermes manages its own model. You set the model in ~/.hermes/config.yaml or through Hermes's own configuration — there is no model field in the Paperclip adapter config. This keeps the "add a Hermes agent" flow one-click: if Hermes is already configured on your machine, you're done.
Session Persistence
When persistSession=true (the default), each run resumes via Hermes's --resume flag. This preserves:
- Conversation context.
- Memories.
- Tool state.
Sessions are tagged as tool source so they don't clutter the user's interactive Hermes history. The adapter's sessionCodec validates and migrates session state between runs.
Skills Integration
The adapter scans two skill sources and merges them in the UI:
- Paperclip-managed skills — bundled with the adapter, togglable from the board UI.
- Hermes-native skills — from
~/.hermes/skills/, read-only, always loaded.
listSkills and syncSkills expose the unified snapshot so the Paperclip UI shows both categories in one view.
Prompt Template Variables
Use {{variable}} syntax in promptTemplate:
| Variable | Description |
|---|---|
{{agentId}} |
Paperclip agent ID |
{{agentName}} |
Agent display name |
{{companyId}} |
Company ID |
{{companyName}} |
Company name |
{{runId}} |
Current heartbeat run ID |
{{taskId}} |
Assigned task/issue ID |
{{taskTitle}} |
Task title |
{{taskBody}} |
Task instructions |
{{projectName}} |
Project name |
{{paperclipApiUrl}} |
Paperclip API base URL |
{{commentId}} |
Comment ID (when woken by a comment) |
{{wakeReason}} |
Reason this run was triggered |
Conditional sections:
{{#taskId}}...{{/taskId}}— included only when a task is assigned.{{#noTask}}...{{/noTask}}— included only on plain heartbeat checks.{{#commentId}}...{{/commentId}}— included only when woken by a comment.
Execution Details
The adapter spawns Hermes Agent in single-query mode (hermes chat -q ...). Hermes processes the task using its full tool suite, then exits. The adapter:
- Captures stdout/stderr and parses token usage, session IDs, and cost.
- Parses raw output into structured
TranscriptEntryobjects (tool cards with status icons). - Post-processes Hermes ASCII banners, setext headings, and
+--+table borders into clean GFM markdown. - Reclassifies benign stderr (MCP init messages, structured logs) so they don't appear as errors.
- Tags sessions as
toolsource. - Reports results back to Paperclip with cost, usage, and session state.
Example
{
"adapterType": "hermes_local",
"adapterConfig": {
"toolsets": "terminal,file,web",
"persistSession": true,
"checkpoints": true,
"timeoutSec": 1800,
"graceSec": 10
}
}