Nm Tome Dig

>- Interactively refine research results by searching deeper into a subtopic or specific channel. Loads the active session and merges new findings into it. U...

Security Scan

The skill's requirements and instructions are consistent with its stated purpose (refining an active research session); it is instruction-only, requests no credentials, and does not attempt unrelated access.

Purpose & Capability

Name/description (interactive refinements of research sessions) match the instructions, which describe loading an active session, running a targeted search, merging results, and updating the report. Access to a session manager and channel filters is proportionate to the stated goal.

Instruction Scope

The SKILL.md is high-level but stays within scope: it directs the agent to load the most recent session, parse a subtopic/channel filter, run targeted searches, merge and re-rank findings, and present results. It does not instruct reading unrelated system files, environment variables, or sending data to external endpoints. The workflow is vague about where searches run (internal agents/channels), which is typical for a plugin that delegates to host services.

Install Mechanism

Instruction-only skill with no install spec and no code files — nothing is written to disk and no external packages or downloads are invoked.

Credentials

The skill declares no required environment variables, credentials, or config paths. The actions it describes (reading and updating an in-agent session) align with that lack of external secrets.

Persistence & Privilege

The skill is not forced-always, does not request elevated persistence, and its described behavior (load/merge session data) is limited to the agent's research sessions rather than modifying other skills or global settings.

Assessment

This is an instruction-only helper meant to refine an existing research session: it will read and update the agent's active session (SessionManager) and invoke targeted searches across channels. It requests no credentials and installs nothing. Before enabling, confirm you are comfortable with a skill having access to your agent's research session data and any channel/search systems the agent uses. If you need specifics about where searches run or what storage is modified, ask the skill author or inspect the host plugin (claude-night-market/tome) for implementation details.

Night Market Skill — ported from claude-night-market/tome. For the full experience with agents, hooks, and commands, install the Claude Code plugin.

Dig Deeper

When To Use

• Drilling into a subtopic after an initial research session
• Narrowing results to a specific channel (e.g. papers only)

When NOT To Use

• Starting a new research topic (use /tome:research first)
• Synthesizing results (use /tome:synthesize)

Refine an active research session interactively.

Workflow

1. Load most recent session via SessionManager
2. Parse the subtopic and optional channel filter
3. Dispatch targeted search (single agent or all channels)
4. Merge new findings into existing session
5. Re-rank and update the saved report
6. Present new findings to user

Error Cases

• No active session: "Start a session first with /tome:research \"topic\""
• Specified channel not in original session: warn and suggest available channels