knowledge-setup

A setup workflow for creating a knowledge base: a directory that contains a wiki and a record of the research projects connected to it.

In plain words
What is it for?
Use it to start a new knowledge base, bootstrap its wiki structure, create its binding record, or re-steer an existing base for a changed topic or audience.
Why use it?
It provides a defined place to store research findings and maintain the link between those findings and the projects that produced them.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/cogni-work/insight-wave/knowledge-setup
Any agent
npx skills add cogni-work/insight-wave --skill knowledge-setup
Clone the repo
git clone --depth 1 https://github.com/cogni-work/insight-wave

Made for: Claude Code, Codex.

Per session 160 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 10,130 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00160 $0.10130
Opus 5 $0.00080 $0.05065
Sonnet 5 $0.00032 $0.02026
Haiku 4.5 $0.00016 $0.01013

Measured 2d ago against content hash 2bbce7fd70cb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

knowledge-setup scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

cogni-knowledge/skills/knowledge-setup/SKILL.md · 454 lines

How it starts

The opening of the file, as written. The whole thing — 454 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Knowledge Setup

Bootstrap a cogni-knowledge knowledge base. A knowledge base is one directory that holds both a cogni-wiki (.cogni-wiki/config.json) and a cogni-knowledge binding manifest (.cogni-knowledge/binding.json). The wiki is the substrate; the binding records which research projects have contributed to it.

This skill is a thin orchestrator. It scaffolds the wiki skeleton natively — the wiki engine is vendored under scripts/vendor/cogni-wiki/, so setup dispatches no cogni-wiki:wiki-setup skill. The cogni-knowledge value-add is the binding manifest plus the one-command workflow.

Read ${CLAUDE_PLUGIN_ROOT}/references/differentiation-thesis.md once at the start of a session to anchor on why wiki-first matters; read ${CLAUDE_PLUGIN_ROOT}/references/delegation-contract.md to remember what cogni-knowledge owns vs. delegates.

When to run

  • User asks to bootstrap, set up, initialize, or start a knowledge base
  • User wants to begin a long-running research area (multiple projects, accumulating findings)
  • User explicitly invokes /cogni-knowledge:knowledge-setup
  • User wants to re-steer an existing base — the domain sharpened, the audience shifted, a new seed theme appeared — via --reframe (re-runs the charter interview against the existing binding; no hand-editing binding.json)

Never run when

  • The target directory already has .cogni-knowledge/binding.json — report the existing binding and stop. Re-initialisation is destructive; surface the existing slug and let the user decide.
  • The user wants a one-off research report — point them at cogni-research:research-setup directly. cogni-knowledge is opinionated about accumulation; cogni-research stays available as a sibling plugin for one-shot reports.

Parameters

Parameter Required Description
--knowledge-slug Yes (prompted) Kebab-case identifier for the knowledge base, e.g. eu-ai-act. Used for the directory name (cogni-knowledge/<slug>/) and the knowledge_slug field in the binding.
--knowledge-title Yes (prompted) Human-readable title, e.g. "EU AI Act knowledge base". Used as the wiki's name in .cogni-wiki/config.json.
--knowledge-root No Override the default knowledge-base directory. Defaults to cogni-knowledge/<knowledge-slug>/ (relative to the current working directory). Both the wiki and the binding live inside this directory.
--description No One-sentence description written into .cogni-wiki/config.json.
--publisher-base-url No Written into .cogni-wiki/config.json as publisher_base_url. Used as last-resort fallback URL when wiki pages have no per-page publisher URL.
--market No Default market for this knowledge base. One of: dach, de, fr, it, pl, nl, es, us, uk, eu. Persisted to binding.json::research_defaults.market; inherited by every knowledge-plan run. Resolved interactively in Step 2.5 when omitted (default dach).
--output-language No Default output language (two-letter code) for this knowledge base. Persisted to binding.json::research_defaults.output_language; inherited by every knowledge-plan run. Resolved interactively in Step 2.5 when omitted — defaults to the chosen market's registry default_output_language (e.g. dachde, frfr, euen).
--prose-density No Default prose density (standard/executive) persisted to binding.json::research_defaults.prose_density. Flag-or-default — not prompted in Step 2.5 (safe default executive).
--tone No Default writing tone persisted to binding.json::research_defaults.tone (see ${CLAUDE_PLUGIN_ROOT}/references/writing-tones.md). Flag-or-default — not prompted (safe default objective).
--citation-format No Default citation format persisted to binding.json::research_defaults.citation_format (ieee/chicago wired; apa/mla/harvard staged). Flag-or-default — not prompted (safe default ieee).
--target-words No Default soft target word count persisted to binding.json::research_defaults.target_words. Flag-or-default — not prompted (safe default 2000).
--charter-domain No One sentence: what this knowledge base is about. Persisted to binding.json::charter.domain (schema 0.1.4). Resolved interactively in Step 2.5 when omitted (engages the charter interview).
--charter-audience No Primary reader of the syntheses this base produces. Persisted to binding.json::charter.audience. Resolved interactively in Step 2.5 when omitted.
--charter-scope No In/out boundaries — geography / segment / horizon, one line. Persisted to binding.json::charter.scope. Resolved interactively in Step 2.5 when omitted.
--open-themes No Pipe-separated seed-theme backlog (e.g. "high-risk systems|conformity assessment|GPAI"). Persisted to binding.json::topic_lineage.open_themes[]; surfaces as the candidate menu for the first research question. Resolved interactively in Step 2.5 when omitted.
--no-charter No Skip the Step 2.5 charter interview AND the Step 5 first-question on-ramp. Use for automation / a flag-only init. The charter fields fall through to "" and open_themes[] to [] (a complete, schema-valid 0.1.4 binding either way).
--no-prelim-search No Keep the Step 2.5 charter interview but skip its optional preliminary scoping scan (stays offline). Same semantics as knowledge-plan --no-prelim-search.
--reframe No Re-steer an existing base instead of bootstrapping a new one. Re-runs the Step 2.5 charter interview against the existing binding and writes the updated charter in place via knowledge-binding.py set-charter (a partial update — only the fields you change). Inverts Step 1's pre-flight: requires an existing binding and aborts when none is found. Skips wiki setup (Steps 2/3), the init call (Step 4), and the Step 5 first-question on-ramp. The charter's data shape (schema 0.1.4) is unchanged.

Read the full file on GitHub · 454 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 454 lines · 160 tokens per session scan A 2bbce7fd70cb

Subscribe to this mod's changes

knowledge-setup is a skill published in the GitHub repository cogni-work/insight-wave (12 stars, last pushed 2d ago), licensed Apache-2.0. It adds 160 tokens to every session and 10,130 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

agent-host-chat-contributions

Build and review cross-cutting agent-host chat behavior through lifecycle contributions. Use when adding turn lifecycle side effects, prompt or context injection, restored-history transformation, protocol-action observation, or when reviewing changes that add code to AgentSideEffects or AgentService.

microsoft/vscode · 56 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens