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/justinjdev/fellowship/gather-lorenpx skills add justinjdev/fellowship --skill gather-loregit clone --depth 1 https://github.com/justinjdev/fellowshipWrote 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/justinjdev/fellowship/gather-lore)<a href="https://agentmods.dev/skills/justinjdev/fellowship/gather-lore"><img src="https://agentmods.dev/badge/skills/justinjdev/fellowship/gather-lore.svg" alt="Measured on agentmods" 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 | $0.00049 | $0.00769 |
| Opus 5 | $0.00024 | $0.00385 |
| Sonnet 5 | $0.00010 | $0.00154 |
| Haiku 4.5 | $0.00005 | $0.00077 |
Grade A, and why
gather-lore 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 3d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gather Lore — Study Patterns Before Writing Code
Overview
Studies existing code to extract the specific patterns and conventions in play. Run this during research — before planning or writing anything — in areas where conventions matter. The patterns you extract here flow into the plan and constrain implementation downstream.
Code generation and deviation checking happen later in the workflow: implementation applies these patterns (quest Phase 3 / TDD), and warden verifies compliance (quest Phase 4).
When to Use
- Entering a part of the codebase you haven't worked in before
- You've had PRs rejected for "wrong approach" in this area
- The task touches patterns you're unsure about (DI, error handling, data access)
Process
Step 1: Find Reference Files
Check CLAUDE.md for a ## Reference Files section. If it exists, read the files listed for the relevant area.
If no reference files are documented, ask the user:
"I need 1-2 examples of files that do something similar to what we're building, that your reviewer would approve of. Can you point me to any?"
If the user can't identify any:
"Let me look at recent merges in this area to find approved patterns." Run:
git log --oneline --diff-filter=A -- [relevant directory] | head -10to find recently added files.
Step 2: Extract Patterns
Read each reference file and produce a structured analysis. Be exhaustive — the patterns you miss are the ones that get flagged in review:
## Patterns in [filename]
### Structure
- [How the file is organized — what comes first, ordering]
- [Import grouping and ordering]
- [Export patterns]
### Dependencies
- [How external dependencies are accessed]
- [How internal dependencies are accessed]
- [Any DI/context patterns]
### Error Handling
- [Error types used]
- [How errors are propagated]
- [How errors are surfaced to callers]
### Data Flow
- [Input validation — where and how]
- [Transformations — where data changes shape]
- [Output — how results are returned]
### Naming
- [Variable naming patterns]
- [Function naming patterns]
- [Type/interface naming patterns]
### Things NOT Done
- [Patterns conspicuously absent — no direct DB calls, no raw HTTP, etc.]
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.
- 3d ago First seen · 89 lines · 49 tokens per session scan A 365deb68e065
gather-lore is a skill published in the GitHub repository justinjdev/fellowship (5 stars, last pushed 20d ago), licensed Apache-2.0. It adds 49 tokens to every session and 769 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 2026-08-31.
Other skills, from other repositories
plan
Use when a request needs shaping before any code is written — a rough or vague prompt to sharpen, an ambiguous idea to design, or a clear-enough task to decompose. One chain-starter that amplifies the prompt, designs the approach, and decomposes it into a batched task file, skipping whichever phases the request…
ai-instruction-detox
AI 指令排毒與規則治理：把散落在 CLAUDE.md、AGENTS.md、skills、context、memory 的規則 原子化、查衝突、去重複、找過時、揪 Prompt Injection，產出可套用的精簡架構與回復方案。 觸發時機：用戶說「指令排毒」「規則太亂」「CLAUDE.md 太長」「規則互相矛盾」「上下文減肥」 「AI 設定治理」「多個 Agent 規則分裂」「context 膨脹」，或要求審查／清理 AI 指令檔。 不要觸發：一般程式重構、產品程式碼審查、單純想縮短一份文件（那是編輯不是治理）。.
team-orchestrator
Spawn an agent team from a predefined role pool (frontend designer, backend developer, researcher, tester, refactorer). Selects the right subset based on the task. Use when the user says "/team", "spawn team", "create team", "start team", or wants to set up a multi-agent development team for a project task.
rapid-domain-mastery
Builds field maps from multi-source academic or technical corpora: expert mental models, debates, assumptions, prerequisites, oral-exam questions, and sprint study plans. Use for exam prep, research onboarding, or rapid topic mastery from multiple sources. Not for plain summaries or single short docs.
continuous-learning
ProductionOS instinct-based learning system. Observes sessions via hooks, extracts patterns, creates atomic instincts with confidence scoring, and auto-promotes high-confidence patterns across projects.
productionos
ProductionOS — dual-target AI engineering operating system for repo-wide audits, upgrade plans, code reviews, strategic product reviews, security sweeps, UX audits, and recursive quality improvement.