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/tayontech/scope/scope-knowledge-loadnpx skills add tayontech/SCOPE --skill scope-knowledge-loadgit clone --depth 1 https://github.com/tayontech/SCOPEWhat 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 | $0.00038 | $0.00663 |
| Opus 5 | $0.00019 | $0.00331 |
| Sonnet 5 | $0.00008 | $0.00133 |
| Haiku 4.5 | $0.00004 | $0.00066 |
Grade A, and why
scope-knowledge-load 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.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SCOPE Knowledge Load
Use this skill at the start of top-level agents before planning, dispatching subagents, writing detections, building controls, or forming investigation hypotheses.
Inputs
The caller provides any known selectors:
ACCOUNT_IDAGENT:scope-audit,scope-controls,scope-exploit, orscope-investigateENTITY: role, user, IP, bucket, key, account, repo, domain, or alert identifier when knownTOPIC: service, TTP, threat actor, alert type, AWS action, or investigation theme when knownTIMEFRAME: the investigation or run window when known
Files To Read
Read these files when present. Missing files are normal on first use.
knowledge/environment.mdknowledge/observations.mdknowledge/coverage-gaps.mdknowledge/exploit-reasoning-notes.mdknowledge/hunt-reasoning-notes.mdconfig/splunk-patterns.md
Selection Rules
Return a bounded KNOWLEDGE_CONTEXT instead of dumping full files.
Prioritize entries that match:
- Exact account ID
- Exact entity
- Same AWS service or log source
- Same TTP, event name, alert type, IOC type, or threat actor
- Recent confirmed or likely-normal observation
- Known coverage gaps that affect the current task
If more than 10 entries match, return the 10 most relevant. Prefer recent, evidence-backed, confirmed entries over stale or needs-review entries.
Interpretation Rules
- Resource identifiers are session-scoped. Do not treat durable knowledge as a source of ARNs, account IDs, bucket names, role names, key IDs, or access key IDs.
- Treat knowledge as context, not ground truth.
- Cite which knowledge entries influenced decisions.
- If live evidence contradicts stored knowledge, trust live evidence and mark the stored knowledge as stale in proposed updates.
- Do not use a knowledge entry as the only evidence for a finding, attack path, detection, remediation, or exploit path.
- Do not broaden scope based only on stored knowledge. Ask the top-level agent to confirm or gather evidence.
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.
- 2d ago First seen · 88 lines · 38 tokens per session scan A 177f19b99bf5
scope-knowledge-load is a skill published in the GitHub repository tayontech/SCOPE (54 stars, last pushed 2mo ago), licensed MIT. It adds 38 tokens to every session and 663 once invoked, about $0.0002 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.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
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…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…