agentleFS
MicrockMicrock/kagi-cli/skills/kagi-assistant/SKILL.md

kagi-assistant

Skill190 starsChanged 8 months agoMITView on GitHub

Use Kagi Assistant for conversational answers, thread continuity, attachments, streaming, model selection, or custom assistants. Use when the task benefits from an account-backed conversation rather than direct search or page extraction.

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use Assistant for conversation mechanics and continuity.

    line 9
  2. Distinctive:

    Delete a thread only when the user asks for it:

    line 39
  3. Distinctive:

    Use human streaming for interactive terminal work.

    line 59
  4. Distinctive:

    Never parse human streaming output as a machine contract.

    line 70

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Kagi Assistant

Use Assistant for conversation mechanics and continuity. Use kagi-ai to choose when to spend AI allowance, and kagi-usage for direct search or page work.

Assistant commands require KAGI_SESSION_TOKEN.

Start a conversation

kagi auth status
kagi assistant "draft a migration checklist" --format markdown
kagi assistant "compare these options" --format toon

State the deliverable, constraints, and audience in the prompt. Never rebuild the output structure in follow-up shell processing when the Assistant can produce it directly.

Continue a thread

kagi assistant --thread-id THREAD_ID "add rollback steps"
kagi assistant thread list
kagi assistant thread get THREAD_ID
kagi assistant thread export THREAD_ID --format markdown

Reuse a thread only when its prior context still belongs to the task. Start a new thread when stale context could bias the answer.

Delete a thread only when the user asks for it:

kagi assistant thread delete THREAD_ID

Attach local context

kagi assistant \
  --attach ./notes.md \
  "turn these notes into a decision record" \
  --format markdown

Attach only files the task needs. Name the file's role in the prompt so the Assistant knows whether it is evidence, a template, or background.

Stream responses

Use human streaming for interactive terminal work. Use structured streaming when another program consumes events.

kagi assistant --stream "explain the tradeoffs"
kagi assistant \
  --stream \
  --stream-output json \
  "produce a release checklist"

Never parse human streaming output as a machine contract.

Custom assistants

kagi assistant custom list
kagi assistant custom get "Researcher"
kagi assistant custom create \
  "CLI Researcher" \
  --web-access \
  --model gpt-5-mini

Inspect an existing custom assistant before changing the workflow around it. Account configuration belongs in kagi-account-config.

Completion criteria

Assistant work is complete when:

  • you used the correct thread or a clean new one;
  • you supplied the required files and constraints;
  • the output format matches its consumer;
  • the final response answers the current prompt, not stale thread context; and
  • you deleted a thread only when the user asked for it.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.