How do I install the Ingest Threads skill?

Run npx skillmds@latest add allanbian1017/ingest-threads in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Ingest Threads safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Ingest Threads?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Ingest Threads free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Ingest Threads?

allanbian1017 (@allanbian1017) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Ingest Threads by allanbian1017 — Agent Skill

Name: Ingest Threads
Author: allanbian1017

Ingest Threads

Fetch a Threads post and produce a summary report in the configured output language (defaulting to English), then mark the Google Task as completed. Use when the user provides a Threads URL to process, says 'process this Threads post', or when invoked by daily-workflow for tasks in the Delegate list.

allanbian1017 c5374a1 2 files · 6.7 KB Updated Sep 17, 2026

File contents

ingest-threads

Full lifecycle for a single Threads task: fetch → summarise → write report → append suggestion → mark done.

Prerequisites: agent-browser installed. For Google Tasks API calls, refer to ../gws-tasks/SKILL.md.

Parameters

Parameter	Required	Description
`THREADS_URL`	Yes	The Threads post URL to process.
`TASK_ID`	Yes	Google Tasks task ID to mark as completed.
`DELEGATE_LIST_ID`	Yes	Google Tasks tasklist ID for the Delegate list.
`SuggestionOutputPath`	Optional	When provided by `daily-workflow`, pass through to `suggestion_log.md` to write to a per-subagent file instead of the default location.

Procedure

Step 1 — Fetch the Threads post content

Follow the full procedure defined in fetch-threads-post:

📄 Read and follow ../fetch-threads-post/SKILL.md

Use a unique session name per task to avoid state collisions:

agent-browser --session ingest-threads-<task-id> open "<THREADS_URL>"
# ... follow fetch-threads-post procedure to expand and extract content ...
agent-browser --session ingest-threads-<task-id> close

Anti-Truncation Standard:

Click "Read more" if visible.
Run agent-browser --session ... scroll down 1000 to trigger lazy-loading.
Use agent-browser --session ... get text body as the primary extraction method.
A complete post ends with a footer, signature, or engagement metrics. Truncated = retry.
The 📄 原始內容 section MUST contain verbatim extraction — never summarize.

Step 2 — Verify and generate a summary in the configured output language

Verification: Compare extracted text length against the visible post. If cut off, repeat extraction.

📄 Read ../content-summary/references/summarise.md

Step 3 — Write the report

📄 Read ../content-summary/references/filename_rules.md

📄 Read ../content-summary/references/output_template.md

Confirm the file is written before proceeding.

Step 4 — Append suggestion to pending backlog

📄 Read ../content-summary/references/ai_analysis.md

📄 Follow ../content-summary/references/suggestion_log.md

{SourceType} = Threads

If SuggestionOutputPath was provided by the caller, pass it through to suggestion_log.md.

Step 5 — Mark the task as completed

gws tasks tasks patch \
  --params '{"tasklist": "<DELEGATE_LIST_ID>", "task": "<TASK_ID>"}' \
  --json '{"status": "completed"}'

Log: "✅ Task '<title>' marked as completed. Report saved to reports/Threads_YYYY_MM_DD/<filename>.md"

Troubleshooting

Threads URL behind login wall: Follow the authenticated access section in fetch-threads-post/SKILL.md.

gws tasks tasks patch fails: Double-check tasklist and task are IDs (not titles).

---name: ingest-threadsdescription: Fetch a Threads post and produce a summary report in the configured output language (defaulting to English), then mark the Google Task as completed. Use when the user provides a Threads URL to process, says 'process this Threads post', or when invoked by daily-workflow for tasks in the Delegate list.---# ingest-threadsFull lifecycle for a single Threads task: fetch → summarise → write report → append suggestion → mark done.> **Prerequisites**: `agent-browser` installed. For Google Tasks API calls, refer to `../gws-tasks/SKILL.md`.---## Parameters| Parameter | Required | Description ||---|---|---|| `THREADS_URL` | Yes | The Threads post URL to process. || `TASK_ID` | Yes | Google Tasks task ID to mark as completed. || `DELEGATE_LIST_ID` | Yes | Google Tasks tasklist ID for the Delegate list. || `SuggestionOutputPath` | Optional | When provided by `daily-workflow`, pass through to `suggestion_log.md` to write to a per-subagent file instead of the default location. |---## Procedure### Step 1 — Fetch the Threads post contentFollow the full procedure defined in `fetch-threads-post`:> 📄 Read and follow `../fetch-threads-post/SKILL.md`Use a unique session name per task to avoid state collisions:```bashagent-browser --session ingest-threads-<task-id> open "<THREADS_URL>"# ... follow fetch-threads-post procedure to expand and extract content ...agent-browser --session ingest-threads-<task-id> close```**Anti-Truncation Standard**:1. Click "Read more" if visible.2. Run `agent-browser --session ... scroll down 1000` to trigger lazy-loading.3. Use `agent-browser --session ... get text body` as the primary extraction method.4. A complete post ends with a footer, signature, or engagement metrics. Truncated = retry.5. The `📄 原始內容` section MUST contain verbatim extraction — never summarize.### Step 2 — Verify and generate a summary in the configured output language**Verification**: Compare extracted text length against the visible post. If cut off, repeat extraction.> 📄 Read `../content-summary/references/summarise.md`### Step 3 — Write the report> 📄 Read `../content-summary/references/filename_rules.md`> 📄 Read `../content-summary/references/output_template.md`Confirm the file is written before proceeding.### Step 4 — Append suggestion to pending backlog> 📄 Read `../content-summary/references/ai_analysis.md`> 📄 Follow `../content-summary/references/suggestion_log.md``{SourceType}` = `Threads`If `SuggestionOutputPath` was provided by the caller, pass it through to `suggestion_log.md`.### Step 5 — Mark the task as completed```bashgws tasks tasks patch \ --params '{"tasklist": "<DELEGATE_LIST_ID>", "task": "<TASK_ID>"}' \ --json '{"status": "completed"}'```Log: `"✅ Task '<title>' marked as completed. Report saved to reports/Threads_YYYY_MM_DD/<filename>.md"`---## Troubleshooting**Threads URL behind login wall**: Follow the authenticated access section in `fetch-threads-post/SKILL.md`.**`gws tasks tasks patch` fails**: Double-check `tasklist` and `task` are IDs (not titles).

Frequently asked questions