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.
npx agentmods add skills/kelp/agent-plugins/kb-capturenpx skills add kelp/agent-plugins --skill kb-capturegit clone --depth 1 https://github.com/kelp/agent-pluginsWhat 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 | $0.00103 | $0.02213 |
| Opus 5 | $0.00051 | $0.01107 |
| Sonnet 5 | $0.00021 | $0.00443 |
| Haiku 4.5 | $0.00010 | $0.00221 |
Grade A, and why
kb-capture 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 yesterday.
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 — 298 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/kb-capture
Capture content into the personal knowledge base as a well-formed wiki note. The command is safe to run from any working directory.
Input
Scope: $ARGUMENTS
- Empty → capture the current conversation as a report.
- Named file or URL → capture that source.
- Topic or title → use it as the note title; synthesize the body from the conversation.
Step 1: Resolve the KB path
Run ${CLAUDE_PLUGIN_ROOT}/scripts/resolve-kb-path.sh and
capture its stdout as kb_path. Every subsequent command
uses "$kb_path" (double-quoted).
kb_path=$(bash "${CLAUDE_PLUGIN_ROOT}/scripts/resolve-kb-path.sh")
If the script exits non-zero, relay its stderr message to the user verbatim and stop — do not guess a fallback path.
Step 2: Classify the note
Pick the note type. When ambiguous, use AskUserQuestion once to clarify — do not guess.
| Type | When to use | Destination |
|---|---|---|
source |
Summary of one external artifact (article, doc, paper) | wiki/sources/external/<category>/<slug>.md |
concept |
Synthesized topic page citing multiple sources | wiki/concepts/<slug>.md |
report |
Deliverable, deep dive, or research output | wiki/reports/<slug>.md |
question |
Unresolved inquiry | wiki/questions/<slug>.md |
Default when $ARGUMENTS is empty: report.
Captured conversations are deliverables.
Source notes require raw_path (a file that already
exists under <kb>/raw/) and url. Do not classify
as source if you are not capturing a specific artifact
with a known raw file.
To capture an external URL as a proper source note (with
a raw_path that exists on disk), run /kb-ingest first;
capturing a URL directly here without an ingested raw file
produces a report, not a source.
The projects/, playbooks/, and dependencies/ wiki
buckets are out of scope for /kb-capture — file notes in
those buckets manually. This does not affect playbook-type
concept notes: those are still captured as concept (see
Step 9's index/playbooks.md entry), since "playbook" is a
tag on a concept note, not a separate capture target.
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.
- yesterday First seen · 298 lines · 103 tokens per session scan A 7e882b68ee3e
kb-capture is a skill published in the GitHub repository kelp/agent-plugins (2 stars, last pushed 9d ago), licensed MIT. It adds 103 tokens to every session and 2,213 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
dummy-dataset
Generate realistic dummy datasets for testing with customizable columns, constraints, and output formats (CSV, JSON, SQL, Python script). Use when creating test data, building mock datasets, or generating sample data for development and demos.
outcome-roadmap
Transform an output-focused roadmap into an outcome-focused one that communicates strategic intent. Rewrites initiatives as outcome statements reflecting user and business impacts. Use when shifting to outcome roadmaps, making a roadmap more strategic, or rewriting feature lists as outcomes.
release-notes
Generate user-facing release notes from tickets, PRDs, or changelogs. Creates clear, engaging summaries organized by category (new features, improvements, fixes). Use when writing release notes, creating changelogs, announcing product updates, or summarizing what shipped.
retro
Facilitate a structured sprint retrospective — what went well, what didn't, and prioritized action items with owners and deadlines. Use when running a retrospective, reflecting on a sprint, creating action items from team feedback, or learning how to run effective retros.
shipping-artifacts
The durable documentation set that makes an AI-built (vibe-coded) app reviewable before shipping. A small core every app needs — architecture, user/permission flows, permissions, variables/secrets, and a test-coverage map — plus conditional docs added only when they apply: emails, scheduled work, SEO, and embedded…
competitive-battlecard
Create sales-ready competitive battlecards comparing your product against a specific competitor — positioning, feature comparison, objection handling, and win/loss patterns. Use when preparing sales teams, creating competitive materials, or responding to 'why not competitor X?'.