# Send HN Daily Brief to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "hn-daily-brief",
    "name": "HN Daily Brief",
    "source": "tencent",
    "type": "skill",
    "category": "内容创作",
    "sourceUrl": "https://clawhub.ai/liu7yong/hn-daily-brief",
    "canonicalUrl": "https://clawhub.ai/liu7yong/hn-daily-brief",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/hn-daily-brief",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=hn-daily-brief",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "SKILL.md",
      "scripts/generate_hn_daily.py",
      "scripts/validate_report.py"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-23T16:43:11.935Z",
      "expiresAt": "2026-04-30T16:43:11.935Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard",
        "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.zip\"",
        "redirectLocation": null,
        "bodySnippet": null
      },
      "scope": "source",
      "summary": "Source download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this source.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/hn-daily-brief"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/hn-daily-brief",
    "downloadUrl": "https://openagent3.xyz/downloads/hn-daily-brief",
    "agentUrl": "https://openagent3.xyz/skills/hn-daily-brief/agent",
    "manifestUrl": "https://openagent3.xyz/skills/hn-daily-brief/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/hn-daily-brief/agent.md"
  }
}
```
## Documentation

### Release Notes

v0.7.0: Add concurrent retry profile for schedule reliability (primary run + immediate retry + delayed retry) with idempotent completion checks to avoid timeout-caused misses.
v0.6.0: Brand rename for distribution as "HN Daily Brief" + full copy cleanup to English-only wording.
v0.5.1: Language cleanup for public distribution (English-first docs; localized markers still supported).
v0.5.0: First public release.

### Parameters

language: output language (default: current user conversation language)
topN: number of items (default: 10)
style: strict | lite (default: strict)
outputDir: output directory (default: /home/ubuntu/.openclaw/workspace/output/hn-daily/)
persist: whether to save file + update index (default: true)
reminderTime: cron time in user timezone, or off

### First-load behavior

On first load (or when user changes params), show effective params once and confirm.
Otherwise, reuse last confirmed params.

### Mandatory execution order

Retry preflight check (required for retry jobs):

If this run is a retry/compensation run, first check whether today's report already exists and is complete (at minimum includes ## Top 10).
If already complete, return NO_REPLY and stop (idempotent exit).


Always fetch fresh inputs for this run (required):

Re-pull current HN Top-N items, article snippets, and comments at run time.
Do not reuse previous report body as source input.


Collect materials via script (data collection only, never user-facing):

scripts/generate_hn_daily.py --style <style> --top <topN> --language <language> --outdir /tmp/hn-daily-draft --materials /tmp/hn-daily-draft/HN-materials.json
Script must only output HN-materials.json (no user-facing report body).
Use HN-materials.json as the only source for final writing.


LLM generation (required, prompt-driven quality):

Use a single strict prompt template to generate final report in selected language.
All content quality constraints (summary depth, comment synthesis style, anti-template wording) must be enforced by prompt, not by script templates.
No “summarize then translate”; generate directly in target language.


Re-check completion before send (required for retry jobs):

If another concurrent run has already persisted a complete report, do not send duplicate content; return NO_REPLY.


Send full final report body to current chat.
If persist=true, write final report to <outputDir>/HN-daily-YYYY-MM-DD.md and update <outputDir>/HN_DAILY_INDEX.md.

### Success criteria (strict)

persist=false: success = full report body delivered in chat.
persist=true: success = chat delivery + file write + index update.
Under concurrent retry profile: success can be achieved by primary run or any retry run; retries must skip if the report is already complete.
If any required condition fails, treat run as failed (do not claim completion).

### Output rules

Send only report body to user (no receipts/status metadata).
Never send script draft markdown directly; user-facing report must come from LLM rewrite over materials JSON.
Output must use one language only: the selected language for the run (no bilingual/mixed-language output).
Real markdown newlines only; never output literal \\n in user-visible content.
Top-N structure per item:

Title
Link
HN link
Heat
Source summary
Comment viewpoint synthesis

### Length rules (by selected language)

Default hard rule: source summary must be >=300 chars in zh (>=200 words in en; equivalent depth for others).
Elastic exception: only when source content is genuinely short/information-limited, summary may be shorter than default target.
Comment viewpoint summary target: zh>=80 chars (equivalent depth for others), with the same short-source exception.
When exception is used, explicitly mark "source is short / info limited" (or equivalent in selected language), prioritize verifiable facts + discussion context + actionable implications, and keep concise but information-dense output.
Never inflate length with generic filler text.

### Comment synthesis rules

Must be in selected language.
Multi-perspective (not binary pro/con).
Per item, output 5 comment viewpoints when available.
If username exists in source comment, it must be preserved in output.
Only use "insufficient comments" when no usable comments are available.
Summarize viewpoints; do not paste long raw quotes.

### Scheduling rules (concurrent retry profile)

If reminderTime != off, create/update a 3-job retry ladder automatically (no duplicates):

Primary run at reminderTime (e.g. HN Daily 06:50)
Immediate retry at reminderTime + 1m (e.g. HN Daily 06:51 Retry)
Delayed retry at reminderTime + 15m (e.g. HN Daily 07:05 Retry)


All retry jobs must be idempotent via completion checks (see Mandatory execution order step 1 and step 5).
Keep job names aligned with schedule and retry role.
Cron payload must be execution-oriented (not passive reminder).
Rationale: cron has no built-in on-failure callback; this retry ladder is the required equivalent for timeout/failure recovery.

### Prompt template requirements (must include)

Generate directly in selected language.
Source summary length must meet selected-language threshold.
Each comment viewpoint summary must meet selected-language threshold.
Summaries must be article-specific and fact-based; avoid reusable boilerplate.
Comment section must synthesize viewpoints (not raw quote dumping), with usernames when available.

### Pre-send quality gate (mandatory)

Validate final report before sending:

zh summary default >=300 chars per item;
zh comment viewpoint default >=80 chars per viewpoint.


Recommended command:

scripts/validate_report.py --report <final_report_path> --language <language>


If an item/viewpoint is below default threshold, it must explicitly include a short-source marker (e.g. source is short / info limited).
If gate fails, rewrite and re-check; do not send failing report.

### Guardrail

If prompt-driven LLM generation path is not available/validated, keep cron disabled until fixed.
If primary + both retries all fail on the same day, send a concise failure alert to the current chat with the first actionable error cause (do not silently drop the run).
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: liu7yong
- Version: 0.7.0
## Source health
- Status: healthy
- Source download looks usable.
- Yavira can redirect you to the upstream package for this source.
- Health scope: source
- Reason: direct_download_ok
- Checked at: 2026-04-23T16:43:11.935Z
- Expires at: 2026-04-30T16:43:11.935Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/hn-daily-brief)
- [Send to Agent page](https://openagent3.xyz/skills/hn-daily-brief/agent)
- [JSON manifest](https://openagent3.xyz/skills/hn-daily-brief/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/hn-daily-brief/agent.md)
- [Download page](https://openagent3.xyz/downloads/hn-daily-brief)