Agent skill

Ccfddl X Posts

by ccfddl in ccfddl/ccf-deadlines

Prepare and, with a verified authenticated publishing route and authorization, publish daily conference-deadline countdown posts for @ccfddl on X/Twitter.

MITAuto-check passedMedia & Creative

Install Ccfddl X Posts

skills CLI
$ npx skills add ccfddl/ccf-deadlines --skill ccfddl-x-posts -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install ccfddl/ccf-deadlines ccfddl-x-posts --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/ccfddl/ccf-deadlines.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/ccfddl-x-posts .claude/skills/ccfddl-x-posts && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
ccfddl-x-posts
GitHub stars
9.4k
Token cost
~4.2k tokens
SKILL.md length
2,193 words
Files
6 (incl. scripts, references)
Skills in repo
3
Repo updated
First seen
Licence
MIT

At a glance

Prepare and, with a verified authenticated publishing route and authorization, publish daily conference-deadline countdown posts for @ccfddl on X/Twitter.

  • Works in 5 steps: Check the connection and authorization → Acquire a coherent current source → Build and inspect → …
  • The recurring CCFDDL campaign
  • SKILL.md covers Campaign contract, 1. Check the connection and…, 2. Acquire a coherent current… and 3. Build and inspect, plus 3 more sections
  • Runs Python scripts from its folder; calls python3; reaches ccfddl.com and x.com

What it does

Ccfddl X Posts is an agent skill from ccfddl/ccf-deadlines. Prepare and, with a verified authenticated publishing route and authorization, publish daily conference-deadline countdown posts for @ccfddl on X/Twitter. Use for the recurring CCFDDL campaign, AI/Data Systems countdown text, email-notification-style social cards, schedule setup, publication checks, or maintaining this workflow. Read the live CCFDDL source, include CCF A and B plus ICLR and MLSys, and link every digest to ccfddl.com.

Its SKILL.md is about 4.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 8 other files, including scripts and reference files (for example `agents/openai.yaml`, `references/config.json` and `references/source-and-publishing.md`).

It sits in Media & Creative, covering Social media graphics and Authorization and RBAC. It works with X (Twitter). The repository describes itself as: ⏰ Agenticly track ccf-ranked & worldwide conference deadlines (Website, Python Cli, Wechat Applet). The licence is MIT.

When your agent uses it

  • The recurring CCFDDL campaign
  • AI/Data Systems countdown text
  • Email-notification-style social cards
  • Publication checks

Example prompts

  • “/ccfddl-x-posts”

Requirements

  • Python 3

Workflow steps

5 steps, taken from the step headings in SKILL.md.

  1. Check the connection and authorization
  2. Acquire a coherent current source
  3. Build and inspect
  4. Publish idempotently
  5. Create or maintain the schedule

What it can do on your machine

Read from SKILL.md and the folder at commit 9e179f4. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 2 files in scripts/ (Python), which the agent can run.

    Shell commands in SKILL.md call:

    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • ccfddl.com
    • x.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Ccfddl X Posts loads about 4.2k tokens when it runs, and up to ~7.7k if it reads all its reference files. Until then it costs about 113 tokens; SKILL.md has 2,193 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~113
When it runs · the whole SKILL.md, loaded when a task matches
~4.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~7.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from ccfddl/ccf-deadlines at commit 9e179f4, republished under its MIT licence (© ccfddl). 2,193 words, ~4,162 tokens.

Download SKILL.mdSave it as .claude/skills/ccfddl-x-posts/SKILL.md (or your agent's skills folder). This skill also uses 5 other files; get the full folder from GitHub.
name
ccfddl-x-posts
description
Prepare and, with a verified authenticated publishing route and authorization, publish daily conference-deadline countdown posts for @ccfddl on X/Twitter. Use for the recurring CCFDDL campaign, AI/Data Systems countdown text, email-notification-style social cards, schedule setup, publication checks, or maintaining this workflow. Read the live CCFDDL source, include CCF A and B plus ICLR and MLSys, and link every digest to ccfddl.com.

CCFDDL X Posts

Create up to two English digests per publication day: one AI, one Data Systems, with concise countdown text, a matching readable PNG, and https://ccfddl.com.

Campaign contract

  • Use exactly the verified X account @ccfddl. Do not treat a display name, brand name, or workspace as account proof.
  • Include CCF A and CCF B conferences plus explicit ICLR and MLSys exceptions. Classify sub: AI as AI and all remaining source subjects as Data Systems. This is the requested campaign grouping, not a claim about the source taxonomy.
  • Publish in the morning in America/Los_Angeles, with automatic daylight-saving changes. Prefer flexible scheduling around 08:00 local; do not impose an exact time unless requested.
  • Use English as the stated operational default. Preserve later user changes in campaign configuration.
  • Publish at most one digest per category per local calendar day. Choose future abstract or paper deadlines within the configured 120-day horizon, up to six text rows. Show only the first three rows in the image, followed by an ellipsis when more eligible rows exist. Keep all candidates when they fit; on overflow, retain representation of both eligible CCF A and CCF B tiers instead of allowing nearer B deadlines to displace every A conference. Record any omissions in the manifest and surface material omissions rather than silently hiding a tier. Skip empty categories.
  • Include no unrelated posts, replies, DMs, follows, or engagement actions. Do not put credentials, account identifiers beyond the public handle, private state, or publication ledgers in the public source repository.
  • Treat this skill as the workflow, not proof that a schedule or publishing connection exists. Never report live posting as enabled without verifying the actual provider and schedule.

Read references/config.json for editable campaign settings. Read references/source-and-publishing.md for source mapping, identity, conflict handling, provider checks, and state transitions.

1. Check the connection and authorization

Before creating a scheduled task, perform a harmless read through the intended publishing route and verify an authenticated X account whose exact normalized handle is ccfddl. Prefer a suitable connected provider; an authorized authenticated X browser is an alternative when no suitable provider is available. For a provider, discover its tools and supported media/publication operations; do not guess them. For a browser route, inspect the actual signed-in profile and composer and verify that the scheduling environment can use that authenticated session. Never assume the current browser login will be available to a future task.

If the chosen route is unconnected, unauthenticated, or points at another account, prepare previews and report the precise blocker. Use the supported plugin or browser login setup flow. Respect cancelled login/setup requests and wait for the user to continue rather than automatically reopening them. Metricool is optional; do not require it or subscribe to a paid service without authorization. For Metricool, a connected plugin alone is insufficient: read brand settings and verify an actual connected X network. Do not guess the info JSON schema or media-file format for createScheduledPost; inspect current tool documentation and official provider documentation first. Do not request or store an API key or password in chat. Installing this skill does not connect X.

For Buffer API/MCP, require a verified public-image hosting route before enabling this image-post campaign: local PNG files cannot be attached directly, and the image URL must remain accessible until publication completes. Never submit a private Library download or signed URL. If this media gate is unresolved, keep publication blocked; do not silently post text alone or claim end-to-end setup is complete. See the Buffer media gate in the source-and-publishing reference.

Use the user's bounded authorization for this recurring campaign. Ask before expanding its audience, topic, account, or scope. Obtain any additional action-specific confirmations required by the platform or applicable policy. Do not publish a preview merely to test the connector.

2. Acquire a coherent current source

Use the verified canonical source: https://github.com/ccfddl/ccf-deadlines, directory conference/. Fetch the latest default-branch commit read-only and pin every YAML file to that same full 40-character SHA. Verify the source against official CFP/date pages whenever a record changes, an ambiguity appears, or a known conflict needs resolution.

Use the bundled snapshot command on a fresh, read-only checkout, or construct the same snapshot with authorized connector results. Do not edit the conference repository as a side effect of publication.

bash
python3 <skill>/scripts/countdown.py snapshot --repo <fresh-read-only-source-checkout> --output <run>/snapshot.json

The snapshot stores repository, commit, fetched_at_utc, complete_catalog, and files (path, exact UTF-8 content, SHA256). When assembling with a connector, set complete_catalog: true only after confirming all conference YAML paths at that commit were retrieved. Sample subsets must have complete_catalog: false and remain previews. snapshot reads committed Git objects, never uncommitted working files.

Never convert TBD, historical patterns, website estimates, conference dates, decisions, rebuttals, or camera-ready dates into submission deadlines. Preserve source seconds. Resolve AoE as UTC−12, fixed UTC offsets explicitly, and IANA zones with date-specific DST. Reject incomplete dates and ambiguous/nonexistent local DST times. Interpret PT as America/Los_Angeles. Quarantine unknown timezone labels rather than guessing.

Apply only the explicitly evidenced, exact-fingerprint EDBT 2027 official clock correction in the config; recheck its official page before publication, retain the original value in the private manifest, and stop for review if the source fingerprint changes to an unrecognized instant. Do not silently publish the known-bad AoE value.

Keep the existing MLSys 2027 exact-clock quarantine until the conflicting official sources agree or the organizer resolves it. Report the discrepancy without selecting whichever time is convenient. Retain MLSys in the campaign scope for other verified editions.

3. Build and inspect

Use a current explicit UTC as-of time. Require Python 3.10+, PyYAML, Pillow, and Liberation Sans (Arial-compatible). Use the executor's existing packages or an authorized reputable package source; do not silently install from an unknown source.

bash
python3 <skill>/scripts/countdown.py build \
  --snapshot <run>/snapshot.json --config <skill>/references/config.json \
  --as-of <current-UTC-ISO8601> --output <run>/posts --preview

For an actual publication run, omit --preview after all gates are satisfied. A non-preview build requires a complete catalog, confirmed editorial configuration, and a snapshot fetched within 15 minutes. Recheck latest source SHA immediately before sending; if it changed, regenerate the affected content before posting.

Inspect every final PNG with an image-view tool. Confirm legible text, no overlaps/clipping, correct category, approved heading and event labels, exact deadline and zone, branding, and no crossed-out footer elements. Keep as-of UTC and source SHA in the private manifest only; retain the canonical link in tweet text, not in the image. Adapt the live repository email reminder template (worker/src/email_reminders.js) for social cards: gray page, white bordered cards, brick-red reminder accent, Arial-compatible sans-serif typography, and countdown → event title → exact deadline hierarchy. Use exactly NAME YEAR Abstract Deadline for abstract events and NAME YEAR Deadline for paper events, retaining a meaningful round label. Do not display an abstract closed warning in tweet text, cards, or alt text; keep the factual flag in the private manifest only. Set the card heading to the exact tweet heading. Remove the in-image website CTA and the visible as-of/source footer; preserve top branding, REMINDER, DEADLINE NOTICE, and the good-luck/team closing. Use the final category-specific plain-text contract below for tweet wording and retain category names on the attached cards. Omit unsubscribe/settings controls and retain ccfddl.com in tweet text only. Preserve precise elapsed-time countdowns rather than copying email calendar-day arithmetic. Every countdown is a snapshot, not a live clock.

Inspect the manifest and each tweet. Honor the skipped-for-review list. Preserve round labels and distinguish abstract from paper. Keep the passed-abstract fact internally, but omit its public qualifier as requested. Such paper deadlines apply to already-registered submissions; do not claim a new submission is still possible. Keep stage_selection: nearest as the default: select the next future deadline per timeline round deterministically. When multiple stages are explicitly selected, support separate rows with stage_selection: all_future; do not change the campaign default merely because the wording changed. Preserve every selected stage and display every abstract_deadline as abstract in both text and card, even when the official site calls it registration. Keep the official term only in source provenance, never as a reg/registration display override.

Keep the full selected digest in the text (up to six rows). Render only its first three rows in the card and add an ellipsis if additional eligible entries exist. Do not add an ellipsis when there are no additional entries. Preserve events (all text rows), card_events (visible image rows), and card_has_more separately in the manifest. Make alt text describe only the visible image rows and the ellipsis, without claiming all text rows are pictured. Keep the default digests concise. Treat the script's weighted-length count and 280-character advisory as editorial diagnostics, not a universal hard limit: the connected account may support Premium long posts. Check the actual provider/account capability. Keep the current approved six-row compact style within the verified account limit; if it does not fit, report the tradeoff instead of silently dropping CCF A entries or changing the requested format. The script counts the canonical URL as 23 characters, NFC-normalizes Unicode, and conservatively overcounts complex emoji. Never use a naive character count, silently truncate, or automatically split a digest into a thread. If the provider returns a text-too-long error, report the exact error and do not rewrite and retry without new authorization. Include alt text from the manifest when the selected route supports it.

Show full SKILL.md (685 more words)Show less
Final plain-text contract

Use exactly CCFDDL deadline reminders (AI) as the AI first line and CCFDDL deadline reminders (Data Systems) as the Data Systems first line, with no trailing colon. Render each selected event as NAME'YY (stage, round N) · N days left, omitting , round N for a single-round conference. Use the existing hour/minute countdown for near-term deadlines. Put one blank line after the heading and one blank line before exactly see details: https://ccfddl.com as the last line. Keep exact dates and timezones on the three visible card rows rather than repeating them in the compact tweet; direct readers to ccfddl.com for the remaining dates. Use ordinary text with no Markdown bold or Unicode bold. Uniformly label abstract_deadline as abstract in the tweet and NAME YEAR Abstract Deadline in its matching card; do not relabel CVPR as reg or registration. Do not invent dates from examples. Omit abstract closed from all public output while retaining its private fact flag. Do not claim that new registrations or new submissions are still possible. The image retains the AI/Data Systems category.

4. Publish idempotently

Use a stable, persistent campaign ledger shared by all runs, or equivalent provider-native durable idempotency. Do not use a temporary directory for live state. If durable state is unavailable, stop before sending.

Compute the publication day in America/Los_Angeles. Check both the ledger and provider's recent/scheduled posts for an existing @ccfddl digest for that day/category. Reconcile existing provider records before proceeding. Never create both a direct post and a separate provider-scheduled copy.

bash
python3 <skill>/scripts/countdown.py ledger check --path <persistent-ledger> --day YYYY-MM-DD --category AI
python3 <skill>/scripts/countdown.py ledger reserve --path <persistent-ledger> --day YYYY-MM-DD --category AI --content-sha256 <manifest-hash>

Prefer the requested Buffer web composer route with native local-PNG upload, after verifying the authenticated account and successful media attachment. Wait for upload completion and inspect the attached image before reserving or publishing. This route is not proven merely because Buffer or OAuth is connected. Do not add a GitHub image-hosting branch or another image host for this campaign.

Only for an explicitly selected Buffer API/MCP fallback, verify the exact final PNG's separately authorized public URL and availability through publication. Without that authorization and verification, keep the API image-post route blocked. Never use private Library or signed download URLs.

Reserve immediately before the single publication request; the reservation uses an exclusive file lock and atomic state replacement. Use the exact approved text, card, and alt text. If upload fails before any post request, determine the exact state without duplicate publication. If sending times out or returns an uncertain result, leave the entry pending, inspect the provider and X, and do not resend automatically. A pending reservation blocks additional sends. Reconcile manually only after establishing whether publication occurred.

After the provider confirms publication, independently read the post and verify the @ccfddl author, text, image, and exact post ID before confirming the ledger:

bash
python3 <skill>/scripts/countdown.py ledger confirm --path <persistent-ledger> --day YYYY-MM-DD --category AI --post-id <verified-numeric-id> --post-url https://x.com/ccfddl/status/<verified-numeric-id>

A queued/scheduled job is not a published tweet; retain its provider job ID in durable campaign state and confirm only after read-back. Report material blockers or conflicts and link verified published posts. Follow the user's notification preferences and avoid repetitive daily success messages.

5. Create or maintain the schedule

Only after the authenticated-route read succeeds, create a daily morning task with the user's supported scheduler. Preserve America/Los_Angeles in the schedule; use flexible timing around 08:00 local. The task must invoke this workflow, refresh source data, perform read-back and idempotency checks, and keep private state outside the public repository. Verify that the task is enabled and tied to the correct provider/account before claiming it is active. Do not create a second schedule if one already exists.

Maintain the reusable skill in the user's personal skill directory. When separately authorized for repository maintenance, mirror only the portable skill files to .agents/skills/ccfddl-x-posts/ through the repository's normal review workflow. Keep the personal installed copy and repository copy aligned when updating behavior. Never commit generated posts, snapshots, private ledger state, credentials, or provider tokens.

Validation

Run the deterministic test suite after changing dates, selection, rendering, or ledger logic:

bash
python3 <skill>/scripts/test_countdown.py

Forward-test a new representative source snapshot, inspect the rendered cards, and exercise the uncertain-send/duplicate guard before enabling changed live behavior. Treat tests and preview creation as validation, never as proof that a post was sent or the schedule is active.

© ccfddl, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 5 other files (scripts, references) in .agents/skills/ccfddl-x-posts of ccfddl/ccf-deadlines.

  • SKILL.md
  • agents/openai.yaml
  • references/config.json
  • references/source-and-publishing.md
  • scripts/countdown.py
  • scripts/test_countdown.py

Open the folder on GitHubat commit 9e179f4

Compare with similar skills

Ccfddl X Posts next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Ccfddl X Posts compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Ccfddl X Posts this skillccfddl/ccf-deadlines9.4k—~4.2kAutomated safety check: PassMIT
Web Asset Generatoralonw0/web-asset-generator5121 repos~6.6kAutomated safety check: PassMIT
Banner CreatorReScienceLab/opc-skills1.8k—~1.3kAutomated safety check: PassApache-2.0
Nuxt SEO Utilsharlan-zw/nuxt-seo-utils130—~2.1kAutomated safety check: PassMIT
Generate Brand Assetscrafter-station/skills111—~1.3kAutomated safety check: PassMIT
Open Graphkostja94/marketing-skills1k—~1.3kAutomated safety check: PassMIT

Similar skills

  • Web Asset Generator

    alonw0/web-asset-generator

    Generate web assets including favicons, app icons (PWA), and social media meta images (Open Graph) for Facebook, Twitter, WhatsApp, and LinkedIn.

    512 GitHub starsUsed in 1 repo~6.6k tokens
    Media & CreativeAuto-check passed
  • Banner Creator

    ReScienceLab/opc-skills

    Create banners using AI image generation. An agent skill from ReScienceLab/opc-skills.

    1.8k GitHub stars~1.3k tokensUpdated yesterday
    Media & CreativeAuto-check passed
  • Nuxt SEO Utils

    harlan-zw/nuxt-seo-utils

    Set up, override, and debug the default SEO head tags in a Nuxt app with the nuxt-seo-utils module.

    130 GitHub stars~2.1k tokensUpdated today
    Media & CreativeAuto-check passed
  • Generate Brand Assets

    crafter-station/skills

    Generate OG images and favicon based on project branding. An agent skill from crafter-station/skills.

    111 GitHub stars~1.3k tokensUpdated 29 days ago
    Media & CreativeAuto-check passed
  • Open Graph

    kostja94/marketing-skills

    When the user wants to add, audit, implement, or optimize Open Graph metadata for social sharing, including og:title, og:description, og:image, og:url, Facebook previews, LinkedIn previews, or…

    1k GitHub stars~1.3k tokensUpdated yesterday
    Media & CreativeAuto-check passed
  • X Posting License Video

    heygen-com/hyperframes-community-skills

    Builds a short animated posting-license card video from an X profile's name, handle, counters and avatar, using a fixed template that the agent only fills in and renders.

    175 GitHub stars~1.2k tokensUpdated 8 days ago
    Media & CreativeAuto-check passed

More from ccfddl/ccf-deadlines

  • Ccfddl Conference Update

    ccfddl/ccf-deadlines

    Add the latest conference edition or newly announced deadline information to ccfddl/ccf-deadlines.

    9.4k GitHub stars~2.9k tokensUpdated today
    Auto-check passed
  • Ccfddl PR Review

    ccfddl/ccf-deadlines

    Review open, non-draft pull requests in ccfddl/ccf-deadlines.

    9.4k GitHub stars~2.4k tokensUpdated today
    Auto-check passed

Works with

Questions about Ccfddl X Posts

What does Ccfddl X Posts do?

Prepare and, with a verified authenticated publishing route and authorization, publish daily conference-deadline countdown posts for @ccfddl on X/Twitter. Ccfddl X Posts is an agent skill from ccfddl/ccf-deadlines. Prepare and, with a verified authenticated publishing route and authorization, publish daily conference-deadline countdown posts for @ccfddl on X/Twitter.

When should I use Ccfddl X Posts?

Ccfddl X Posts fits situations like: the recurring CCFDDL campaign; AI/Data Systems countdown text; email-notification-style social cards; publication checks.

How do I install Ccfddl X Posts in Claude Code?

Run `npx skills add ccfddl/ccf-deadlines --skill ccfddl-x-posts -a claude-code`. Or copy the skill folder (.agents/skills/ccfddl-x-posts in ccfddl/ccf-deadlines) into .claude/skills/ccfddl-x-posts in your project. Claude Code loads it when a task matches its description.

How do I install Ccfddl X Posts in Codex?

Run `npx skills add ccfddl/ccf-deadlines --skill ccfddl-x-posts -a codex`. Or copy the skill folder (.agents/skills/ccfddl-x-posts in ccfddl/ccf-deadlines) into .agents/skills/ccfddl-x-posts in your project. Codex loads it when a task matches its description.

Can I use Ccfddl X Posts in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ccfddl/ccf-deadlines --skill ccfddl-x-posts -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ccfddl-x-posts, .gemini/skills/ccfddl-x-posts, .github/skills/ccfddl-x-posts and .opencode/skills/ccfddl-x-posts in your project.

What does Ccfddl X Posts need to run?

Going by SKILL.md and its folder, Ccfddl X Posts needs Python for the scripts in its folder and the command-line tools its instructions call (python3). Our summary lists: Python 3.

Does Ccfddl X Posts access the network?

SKILL.md names 2 domains. In commands or code: ccfddl.com and x.com; the agent is likely to contact these when it follows the instructions. This is read from the text; nothing was executed.

Is Ccfddl X Posts safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Ccfddl X Posts use?

Ccfddl X Posts is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Ccfddl X Posts use?

About 4.2k tokens (SKILL.md is roughly 17k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 3.6k tokens, read only when the agent opens those files.

What are the alternatives to Ccfddl X Posts?

Skills that share tags, products or a category with Ccfddl X Posts: Web Asset Generator (alonw0/web-asset-generator, 512 stars), Banner Creator (ReScienceLab/opc-skills, 1.8k stars), Nuxt SEO Utils (harlan-zw/nuxt-seo-utils, 130 stars) and Generate Brand Assets (crafter-station/skills, 111 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Ccfddl X Posts?

ccfddl (a GitHub organization) maintains it in ccfddl/ccf-deadlines, which has 9,413 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on October 7, 2026.

Source: ccfddl/ccf-deadlines on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.