Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add cogni-work/insight-wave/plugin install cogni-knowledgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/cogni-work/insight-wave/knowledge-fetch)<a href="https://agentmods.dev/skills/cogni-work/insight-wave/knowledge-fetch"><img src="https://agentmods.dev/badge/skills/cogni-work/insight-wave/knowledge-fetch/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/cogni-work/insight-wave/knowledge-fetch"><img src="https://agentmods.dev/badge/skills/cogni-work/insight-wave/knowledge-fetch.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00166 | $0.04272 |
| Opus 5.5 | $0.00066 | $0.01709 |
| Sonnet 5.5 | $0.00033 | $0.00854 |
| Haiku 4.5 | $0.00017 | $0.00427 |
Grade A, and why
knowledge-fetch 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 179 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge Fetch
Phase 3 of the inverted pipeline. The WebFetch body-pull happens inside the Phase-2 curators, so this skill no longer fetches by default — it reads each candidate's fetch sub-object from <project>/.metadata/candidates.json and assembles the canonical <project>/.metadata/fetch-manifest.json directly. Its remaining active job is opt-in cobrowse (--cobrowse or the interactive prompt): it walks the user through enabling the Claude-in-Chrome browser extension and dispatches source-fetcher sequentially to do two things — recover WebFetch misses (merging rescues into the manifest) and additively top up thin primary-tier sources (reading the fuller browser body beyond WebFetch's cap, superseding the cached body only if strictly longer, never degrading a usable one).
Read ${CLAUDE_PLUGIN_ROOT}/references/inverted-pipeline.md §"Phase 3 — knowledge-fetch" and references/fetch-cache-design.md once to anchor on the contract.
When to run
candidates.jsonexists for the project (Phase 2 has run) AND eitherfetch-manifest.jsondoes not yet exist OR the user explicitly wants to re-fetch (e.g., after evicting stale cache entries)- User explicitly invokes
/cogni-knowledge:knowledge-fetch
Never run when
- No
candidates.jsonexists at<project_path>/.metadata/— offerknowledge-curatefirst. - No
binding.jsonexists at the resolved knowledge root.
Parameters
| Parameter | Required | Description |
|---|---|---|
--knowledge-slug |
Yes | Slug of the bound knowledge base. |
--project-path |
Yes | Absolute path to the project directory. |
--knowledge-root |
No | Override the default knowledge-base directory. |
--max-age-days |
No | Cache freshness window in days. Default: from binding.curator_defaults.fetch_cache_max_age_days (30). Forwarded to the cobrowse source-fetcher. |
--cobrowse |
No | Opt in to browser-assisted cobrowse — both recovery of WebFetch misses and additive top-up of thin primary-tier sources (reading the fuller body beyond WebFetch's cap). Default OFF (autonomous runs stay browser-free). |
--no-cobrowse |
No | Force cobrowse off (both recovery and top-up) even in an interactive session (suppresses the prompt). |
--batch-size |
No | Vestigial — the default WebFetch path no longer dispatches per batch (the curators already fetched). Cobrowse is per-URL sequential through the single shared browser tab. Kept for backward-compat; ignored. |
--tier |
No | Restrict the cobrowse sets (both misses and top-up candidates) to a single tier (primary, secondary, supporting). Default: all tiers (the top-up set is already primary-only by construction). No longer bounds WebFetch cost — that is the curator's max_candidates_per_sq (Phase 2). |
--dry-run |
No | Print the manifest plan (fetched / unavailable counts from the candidates' fetch sub-objects, cobrowse-eligible miss count, and cobrowse top-up candidate count) without dispatching cobrowse. |
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.
- 1mo ago Changed · -14 lines ea875313688b
- 1mo ago First seen · 193 lines · 166 tokens per session scan A b333b9ecd756
knowledge-fetch is a skill published in the GitHub repository cogni-work/insight-wave (13 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 166 tokens to every session and 4,272 once invoked, about $0.0007 per session on Opus 5.5. 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.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
interactive-login
Log a CLI in with a browser/device-code flow (aws sso, gh, glab, gcloud, …) and save the result to the keychain so later commands can request it through execute.credentials.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
azure-messaging-webpubsub-java
Build real-time web applications with Azure Web PubSub SDK for Java. Use when implementing WebSocket-based messaging, live updates, chat applications, or server-to-client push notifications.
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
open-design-browser
Inspect and exercise local HTML design artifacts and library prototype previews with the existing agent-browser CLI, including screenshots, interaction checks, and rendered DOM snapshots. Use when rendered evidence helps author or review a design; excludes browser installation, remote services, and unrelated browser…