跳到正文
FunCoding

搜索

搜索文档、Skill 和 MCP

telegram-reader

Read Telegram channels and groups for financial news and market research with the tdl CLI: list chats and channels, find a channel by name, and export recent messages or a date or ID range. Use this skill whenever the user wants to check their Telegram, see what's new in their channels, read or export messages from a news, crypto, or trading channel or group, or gather market news and signals posted on Telegram. Read-only: it cannot send messages or join or leave channels.

科研3.4kplugins/social-readers/skills/telegram-reader/SKILL.md

安装

把这段话发给 Claude Code、Codex 或 Cursor。智能体会先检查安全性,你确认后才安装。

读取 https://funcoding.ai/skills/himself65/finance-skills/telegram-reader/install.md ,按里面的步骤帮我安装这个 Skill。

SKILL.md

Telegram News Skill (Read-Only)

Reads Telegram channels and groups for financial news and market research using tdl, a Telegram CLI tool.

This skill is read-only. It is designed for financial research: reading channel messages, monitoring financial news channels, and exporting message history. It does NOT support sending messages, joining/leaving channels, or any write operations.


Step 1: Ensure tdl Is Installed

Current environment status:

!`(command -v tdl && tdl version 2>&1 | head -3 || echo "TDL_NOT_INSTALLED") 2>/dev/null`

If the status above shows a version number, tdl is installed — skip to Step 2.

If TDL_NOT_INSTALLED, install tdl based on the user's platform:

PlatformInstall Command
macOS / Linuxcurl -sSL https://docs.iyear.me/tdl/install.sh | sudo bash
macOS (Homebrew)brew install telegram-downloader
Linux (Termux)pkg install tdl
Linux (AUR)yay -S tdl
Linux (Nix)nix-env -iA nixos.tdl
Go (any platform)go install github.com/iyear/tdl@latest

Default to Homebrew on macOS. On Linux the install script runs with sudo, so confirm with the user before using it, or offer their package manager instead.


Step 2: Ensure tdl Is Authenticated

Current auth status:

!`(tdl chat ls --limit 1 2>&1 >/dev/null && echo "AUTH_OK" || echo "AUTH_NEEDED") 2>/dev/null`

If AUTH_OK, skip to Step 3.

If AUTH_NEEDED, guide the user through login. Login requires interactive input — the user must enter their phone number and verification code manually.

Login methods

Method A: QR Code (recommended — fastest)

tdl login -T qr

A QR code will be displayed in the terminal. The user scans it with their Telegram mobile app (Settings > Devices > Link Desktop Device).

Method B: Phone + Code

tdl login -T code

The user enters their phone number, then the verification code sent to their Telegram app.

Method C: Import from Telegram Desktop

If the user has Telegram Desktop installed and logged in:

tdl login

This imports the session from the existing desktop client. The desktop client must be from the official website, NOT from the App Store or Microsoft Store.

Namespaces

By default, tdl uses a default namespace. To manage multiple accounts:

tdl login -n work -T qr      # Login to "work" namespace
tdl chat ls -n work           # Use "work" namespace for commands

Important login notes

  • Login is a one-time operation. The session persists on disk after successful login.
  • If login fails, ask the user to check their internet connection and try again.
  • Never ask for or handle Telegram passwords/2FA codes programmatically — always let the user enter them interactively.

Step 3: Identify What the User Needs

Match the user's request to one of the read operations below.

User RequestCommandKey Flags
List all chats/channelstdl chat ls-o json, -f "FILTER"
List only channelstdl chat ls -f "Type contains 'channel'"-o json
Export recent messagestdl chat export -c CHAT -T last -i N--all, --with-content
Export messages by time rangetdl chat export -c CHAT -T time -i START,END--all, --with-content
Export messages by ID rangetdl chat export -c CHAT -T id -i FROM,TO--all, --with-content
Export from a topic/threadtdl chat export -c CHAT --topic TOPIC_ID--all, --with-content
Search for a channel by nametdl chat ls -f "VisibleName contains 'NAME'"-o json

Chat identifiers

The -c flag accepts multiple formats:

FormatExample
Username (with @)-c @channel_name
Username (without @)-c channel_name
Numeric chat ID-c 123456789
Public link-c https://t.me/channel_name
Phone number-c "+1 123456789"
Saved Messages-c "" (empty)

Step 4: Execute the Command

Listing chats

# List all chats
tdl chat ls

# JSON output for processing
tdl chat ls -o json

# Filter for channels only
tdl chat ls -f "Type contains 'channel'"

# Search by name
tdl chat ls -f "VisibleName contains 'Bloomberg'"

Exporting messages

Always use --all --with-content to get text messages (not just media):

# Last 20 messages from a channel
tdl chat export -c @channel_name -T last -i 20 --all --with-content -o /tmp/tdl-export.json

# Messages from a time range (Unix timestamps)
tdl chat export -c @channel_name -T time -i 1710288000,1710374400 --all --with-content -o /tmp/tdl-export.json

# Messages by ID range
tdl chat export -c @channel_name -T id -i 100,200 --all --with-content -o /tmp/tdl-export.json

Key rules

  1. Check auth first — run tdl chat ls --limit 1 before other commands to verify the session is valid
  2. Always use --all --with-content when exporting messages for reading — without these flags, tdl only exports media messages
  3. Use -o FILE to save exports to a file, then read the JSON — this is more reliable than parsing stdout
  4. Start with small exports — use -T last -i 20 unless the user asks for more
  5. Use filters on chat ls to help users find the right channel before exporting
  6. Read-only — don't send messages, join or leave channels, or modify anything
  7. Convert timestamps — when the user gives dates, convert to Unix timestamps for the -T time filter

Working with exported JSON

After exporting, read the JSON file and extract the relevant information:

# Export messages
tdl chat export -c @channel_name -T last -i 20 --all --with-content -o /tmp/tdl-export.json

# Read and process the export
cat /tmp/tdl-export.json

The export JSON contains message objects with fields like id, date, message (text content), from_id, views, and media metadata.


Step 5: Present the Results

Lead with the most relevant news or market updates rather than a raw dump. Give each message its timestamp and source channel, group messages from several channels by theme (macro, earnings, crypto), and call out breaking news, price targets, and earnings surprises. For channel lists, show name, member count, and type.


Step 6: Diagnostics

If something isn't working:

ErrorCauseFix
not authorized or session errorsNot logged in or session expiredRun tdl login -T qr to re-authenticate
FLOOD_WAIT_XRate limited by TelegramWait X seconds, then retry
CHANNEL_PRIVATENo access to channelUser must join the channel in their Telegram app first
tdl: command not foundtdl not installedInstall using Step 1

Reference Files

  • references/commands.md — Complete tdl command reference for reading channels and exporting messages

Read the reference file when you need exact command syntax or detailed flag documentation.

相似的 Skill

lead-research-assistant
ComposioHQ/awesome-claude-skills77k

lead-research-assistant

Identifies high-quality leads for your product or service by analyzing your business, searching for target companies, and providing actionable contact strategies. Perfect for sales, business development, and marketing professionals.

科研

13c-metabolic-flux
K-Dense-AI/scientific-agent-skills48k

13c-metabolic-flux

Estimates intracellular metabolic fluxes from steady-state carbon-13 isotope-tracing measurements using validated atom maps, mfapy isotope simulation, constrained multistart fitting, and flux-profile diagnostics. Use for 13C-MFA, carbon tracing, mass isotopomer distributions (MDVs/MIDs), positional isotopomers, parallel tracer experiments, and determining whether labeling data constrain a pathway flux. Distinguishes measured-label inference from COBRA flux balance analysis and flags experiments requiring nonstationary MFA.

科研

datamol
K-Dense-AI/scientific-agent-skills48k

datamol

Pythonic wrapper around RDKit with simplified interface and sensible defaults. Preferred for standard drug discovery including SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformers, parallel processing. Returns native rdkit.Chem.Mol objects. For advanced control or custom parameters, use rdkit directly.

科研

biopython
K-Dense-AI/scientific-agent-skills48k

biopython

Provides Biopython workflows for sequence manipulation, file parsing (FASTA/GenBank/PDB), phylogenetics, and programmatic NCBI/PubMed access (Bio.Entrez). Supports batch processing, custom molecular-biology pipelines, BLAST automation, structure analysis, and motif analysis.

科研

bulk-rnaseq
K-Dense-AI/scientific-agent-skills48k

bulk-rnaseq

Prepares bulk RNA-seq FASTQ, Salmon, STAR or featureCounts output for gene-level differential expression. Covers nf-core/rnaseq and standalone quantification, biological replication, strandedness, reference provenance, validated count assembly and a PyDESeq2 handoff. Use for FASTQ-to-counts analysis, nf-core/rnaseq configuration, STAR/Salmon quantification, or building a counts matrix for DESeq2. For single-cell data use scanpy; for statistical fitting alone use pydeseq2.

科研

alphagenome
K-Dense-AI/scientific-agent-skills48k

alphagenome

Looks up precomputed AlphaGenome Atlas effects for any GRCh38 single-nucleotide variant (AVI score with Phred and 18 SHAP feature attributions, plus raw and quantile scores for RNA-seq, DNase, ATAC, ChIP-TF, ChIP-histone, CAGE, PRO-cap, splicing, polyadenylation and contact-map tracks), scores variants or scans windows on demand with the AlphaGenome model for human and mouse (variant scoring, in silico mutagenesis, REF-versus-ALT track prediction), and builds Atlas website deep links. Use when the user mentions AlphaGenome, AlphaGenome Atlas, AVI or AlphaGenome Variant Impact, DeepMind variant effect prediction, or wants to prioritise or mechanistically interpret non-coding, regulatory, splicing, enhancer, promoter, or chromatin-accessibility effects of SNVs from a VCF, credible set, or region. Research use only; not a clinical tool.

科研