Telegram Context
PassAudited by ClawScan on May 1, 2026.
Overview
This instruction-only skill matches its stated purpose, but when enabled it reads recent Telegram chat messages and sends them into the AI context, so it should be used carefully in sensitive chats.
This skill appears purpose-aligned and transparent. Before installing, be comfortable with recent Telegram messages from the current chat being read into the AI context and sent to your configured model provider. Prefer manual fetch or a low fetch count for sensitive chats, and turn the skill off when continuity is not needed.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent can read recent messages from the current Telegram chat when the skill is used.
The skill declares message-read access, which is necessary for fetching Telegram history but still gives the agent access to chat content.
permissions:
- message:readOnly enable it in chats where reading recent history is acceptable, and verify the Telegram gateway permissions match your expectations.
Private or sensitive Telegram messages may be exposed to the configured AI provider and possibly session logs.
The artifact clearly states that fetched Telegram messages are forwarded into the model context and to the configured provider.
Message content is **included in the agent's context window** and sent to the configured LLM provider
Use manual fetch or a low fetch count for sensitive conversations, and disable the skill when you do not want chat history sent to the model.
Old or third-party Telegram messages could influence the agent’s responses if they are not clearly treated as untrusted history.
Recent Telegram messages are inserted into the agent’s working context; prior chat messages may include instructions or content that should be treated as historical context rather than authoritative commands.
Provide history as context for the conversation
Treat fetched messages as background context only, and require fresh user confirmation before the agent takes important actions based on prior chat content.
Once enabled, the skill may continue fetching recent Telegram history at future session starts until turned off.
The skill stores an enabled/disabled state and fetch settings, allowing automatic history fetching to persist across sessions when enabled.
State is stored in `memory/telegram-context.json`
Use `/telegram-context status` to check the setting and `/telegram-context off` before sensitive conversations.