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/curt-park/autology/capture-knowledgenpx skills add Curt-Park/autology --skill capture-knowledgegit clone --depth 1 https://github.com/Curt-Park/autologyWrote 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/curt-park/autology/capture-knowledge)<a href="https://agentmods.dev/skills/curt-park/autology/capture-knowledge"><img src="https://agentmods.dev/badge/skills/curt-park/autology/capture-knowledge.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 | $0.00084 | $0.01001 |
| Opus 5 | $0.00042 | $0.00500 |
| Sonnet 5 | $0.00017 | $0.00200 |
| Haiku 4.5 | $0.00008 | $0.00100 |
Grade A, and why
capture-knowledge 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 6d 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Overview
Capture knowledge from conversation context into docs/ as markdown nodes. Save immediately — asking confirmation interrupts flow and discourages capture.
Precondition: if triage has not run this session, run /autology:triage-knowledge first.
What Capture Targets
Capture the items triage classified as new (→ capture):
- Decisions made (technology choices, architectural choices)
- Components created or modified
- Conventions or patterns established
- Concepts or domain knowledge explained
- Issues or technical debt identified
When NOT to capture:
- Session-specific context (current task, temporary state)
- Incomplete or unverified information
- Items triage classified as existing (→ sync, not capture)
Quick Reference
| Type | Signals |
|---|---|
| decision | "chose", "decided", "selected", "adopted" |
| component | "created", "built", "implemented", new service/module |
| convention | "always", "never", "must", "should", "the rule is" |
| concept | lifecycle, workflow, domain model, process |
| pattern | reusable approach, strategy, pattern |
| issue | bug, technical debt, bottleneck, known problem |
| session | work session summary, "finished", "completed" |
type = primary classification (what kind?); tags = cross-cutting topics (what about?).
Process
1. Receive Triage Output
Use triage's new items list as the capture scope:
New items from triage (→ capture):
- [item description] — no matching node
Suggested relations: [[foo]], [[bar]] (shared tags: architecture)
- [item description] — no matching node
Suggested relations: none
2. Decide node granularity
Not every triage item needs its own file. Before creating, ask: does this item have enough standalone substance to be useful on its own, or is it a detail of something larger?
- Own node: has a distinct title, multiple meaningful sentences, likely to be linked or searched independently
- Decisions always get their own node — even if brief. A decision is "chose X over Y because Z": the rationale makes it independently searchable and worth linking from other contexts. E.g., "chose JWT over session cookies because we need a stateless API" →
type: decisionnode, not a footnote in the component doc. - Fold into parent: a pure behavior detail, edge case, or implementation consequence of another item in the same batch (e.g., "returns 401 on invalid token" is a consequence of the JWT middleware, not a choice made with rationale)
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 6d ago First seen · 100 lines · 84 tokens per session scan A 362727400b9e
capture-knowledge is a skill published in the GitHub repository Curt-Park/autology (24 stars, last pushed 6mo ago), licensed MIT. It adds 84 tokens to every session and 1,001 once invoked, about $0.0004 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
mama-context
Always-on background context injection from MAMA memory. Automatically surfaces relevant decisions when you work on code, without explicit invocation.
mama-install
Install MAMA OS and follow its self-teaching CLI until the first report is confirmed.
vault
Search, read, write, and manage files in the Kept conversation vault. Use when the user asks about past conversations, wants to save notes, needs to find specific content, or wants to organize their vault.
kb-session-review
Review this session's knowledge base usage. Use when the user asks whether the KB was updated, what was captured, if anything was missed, or wants a session retrospective on knowledge capture quality.
wiki-retrieve
Build and query a vault-local contextual BM25 retrieval index with optional multilingual Nomic cosine reranking; use for retrieve, hybrid retrieval, BM25, rerank, contextual retrieval, chunk search, vault search, semantic search, find relevant passages, or retrieval diagnostics. Derived caches stay under .vault-meta…
memory-audit-pattern-extraction
模式提取与失效解药分析。当发现多条记忆在讲同一个教训,或发现自己在一而再再而三地犯同样的错误时使用。.