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 skills add WellApp-ai/Well --skill decision-capturegit clone --depth 1 https://github.com/WellApp-ai/WellWrote 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/wellapp-ai/well/decision-capture)<a href="https://agentmods.dev/skills/wellapp-ai/well/decision-capture"><img src="https://agentmods.dev/badge/skills/wellapp-ai/well/decision-capture.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 5.1 | $0.00018 | $0.01941 |
| Opus 5 | $0.00009 | $0.00971 |
| Sonnet 5 | $0.00004 | $0.00388 |
| Haiku 4.5 | $0.00002 | $0.00194 |
Grade A, and why
decision-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 7d 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 — 302 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Decision Capture Skill
Lightweight skill for capturing the "patine" - the accumulated wisdom of why decisions were made and alternatives rejected. Triggered at Gates when human provides KO or challenges a proposal.
When to Use
- At any Gate when human provides KO
- When human challenges or rejects a proposal
- When debug skill finds a pattern worth remembering
- When significant technical decision is made
Trigger Conditions
| Trigger | Context | Type |
|---|---|---|
| Gate 1 KO | Wireframe rejected | Decision |
| Gate 2 KO | Scope option rejected | Decision |
| Gate 3 KO/BLOCK | Phasing challenged | Decision |
| Gate 4 KO | Technical approach rejected | Decision |
| Gate 5 Changes | PR review feedback | Decision |
| Debug Pattern | Recurring issue found | Kaizen |
| DIG 3+ times | Same wireframe refined repeatedly | Kaizen |
| Same error fixed 2+ times | Fix pattern emerged | Kaizen |
| User corrects assumption | AI was wrong about something | Hansei |
| Takt warning exceeded | Phase took longer than target | Hansei |
| Jidoka Tier 2/3 | Escalation to human required | Hansei |
Phase 1: Detect Decision Type
Categorize the decision:
| Type | Signal | Example |
|---|---|---|
| Technical | Code, architecture, library | "Don't use GraphQL subscriptions" |
| UX | Interaction, visual, flow | "Sidebar navigation, not top nav" |
| Process | Workflow, phasing, priority | "Ship auth before tables" |
Phase 2: Prompt for Rationale
Ask for brief rationale (keep it light):
I'll note this decision for future reference.
**In one sentence, why this decision?**
Examples:
- "We tried X in 2024, broke production"
- "Users missed this in testing"
- "Conflicts with our caching strategy"
(Press Enter to skip if you prefer not to explain)
If human declines: Record decision without rationale (still valuable).
Phase 3: Attribute
Capture metadata:
| Field | Source |
|---|---|
| Who | Current user (from context) |
| When | Current date |
| Domain | From branch name or changed files |
| Gate | Which Gate triggered capture |
| Related Task | Notion task ID if available |
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.
- 7d ago First seen · 302 lines · 18 tokens per session scan A 1cd931aa6fff
decision-capture is a skill published in the GitHub repository WellApp-ai/Well (340 stars, last pushed 1mo ago), licensed MIT. It adds 18 tokens to every session and 1,941 once invoked, about $0.0001 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
agent-docs
Use when writing documentation optimized for AI agent consumption - SKILL.md files, README files, API docs, or any documentation that will be read by LLMs in context windows.
cloudflare-router
Use when manage Cloudflare DNS, CDN, and security rules via API. Use when configuring domains, SSL, WAF, or edge caching.
dify-workflow
Use when dify AI workflow platform — LLM apps, knowledge bases, agents, workflow orchestration, API deployment. Use when working with dify workflow.
llm-deployment
Use when lLM deployment and serving — vLLM, Ollama, TGI, llama.cpp. Model quantization, GPU optimization, API serving. Use when working with llm deployment.
geminiignore-finops
Configure and optimize .geminiignore files for AI context window efficiency and token cost reduction (FinOps).
hive.note-taking
Maintain a free-form scratchpad of decisions, extracted values, and open questions so context pruning doesn't lose anything you still need.