A skill for capturing durable lessons, decisions, conventions, and other knowledge from a work session.

Use it at the end of a debugging or development session to route lasting knowledge to shared project instructions, specifications, or other appropriate stores.

It prevents useful discoveries from disappearing while filtering out temporary details that would not help in future sessions. Writes are proposed for the user's approval.

compound — Skill for Claude Code and Codex by toverux/cantrips

Install

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.

Any agent

npx skills add toverux/cantrips --skill compound

Clone the repo

git clone --depth 1 https://github.com/toverux/cantrips

Made for: Claude Code, Codex.

Or install cantrips, the plugin that ships this one along with the rest of its 25 skills, 3 agents.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/toverux/cantrips/compound/github.svg)](https://agentmods.dev/skills/toverux/cantrips/compound)

Your own site

<a href="https://agentmods.dev/skills/toverux/cantrips/compound"><img src="https://agentmods.dev/badge/skills/toverux/cantrips/compound/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/toverux/cantrips/compound"><img src="https://agentmods.dev/badge/skills/toverux/cantrips/compound.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 78 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,369 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00078	$0.01369
Opus 5.5	$0.00031	$0.00548
Sonnet 5.5	$0.00016	$0.00274
Haiku 4.5	$0.00008	$0.00137

Measured 13d ago against content hash 22d094147f3e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

compound 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 13d 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.

skills/compound/SKILL.md · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Harvest what this session learned so future sessions inherit it. Only knowledge that compounds gets written; every write is approved first, by the user or by a deputy engaged in their place.

1. Scan the session

Sweep the whole session for candidates: corrections the user made and preferences they stated, conventions that emerged, durable decisions settled or reversed, gotchas and root causes uncovered, approaches that failed and why, domain terms coined or clarified ("we call this X, not Y"), and workflow friction a procedure could remove.

2. Apply the quality bar

For each candidate, two questions: would this change a future agent's behavior in a different session, and is it non-obvious and stable? Any "no" → discard.

This bar is the defence against the eager-auto-memory failure mode: session-specific trivia — in-flight task state, one-off values, things any agent would infer anyway — must die here.

If nothing clears the bar, say so in one line and stop.

3. Route each survivor

Project AGENTS.md — durable shared preferences and conventions, cheap enough to always load. Domain terms go in its glossary section; when the glossary outgrows AGENTS.md, propose graduating it to a CONCEPTS.md file.
docs/adr/ (opt-in store) — durable decisions: a project-shaping choice settled among live alternatives — where AGENTS.md holds preferences and docs/solutions/ holds problem-shaped learnings. One record per decision, per adr-format.md. Reversing a standing decision routes as a pair — a new ADR plus the status flip on the superseded record.
docs/solutions/ (opt-in store) — problem-shaped learnings: root cause, gotcha, what didn't work. One small file per solved problem, per solutions-format.md. Expensive knowledge, loaded on demand.
Project-local rules — path-scoped project conventions, when the project uses rules files.
Skills — project-specific procedures → the project's skills directory (.claude/skills/ on Claude Code, or the harness's equivalent); generic workflow improvements → the user's personal skills collection, if configured.
User-global memory file (~/.claude/CLAUDE.md on Claude Code, ~/.codex/AGENTS.md on Codex CLI, or the harness's equivalent) — personal-only preferences, and the staging area to trial a candidate durable preference before graduating it to a shared file.

Read the full file on GitHub · 61 lines

Files

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.

Changes

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.

13d ago Changed · +2 lines · +4 tokens per session 22d094147f3e
1mo ago Changed 0407d13a9239
1mo ago Changed · +4 lines ca67675deaf2
1mo ago First seen · 55 lines · 74 tokens per session scan A 9c7e1f60e953

Subscribe to this mod's changes

compound is a skill published in the GitHub repository toverux/cantrips (3 stars, last pushed yesterday), licensed MIT. It adds 78 tokens to every session and 1,369 once invoked, about $0.0003 per session on Opus 5.5. 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.

Claude Code Code review Memory and context agent-skills ai-agents claude-plugins codex-cli

context-canvas-checkpoint

Maintain an optional session task map and explicit retrievable references in Context Canvas Codex. Use when navigation or long-context offload would materially help with goals, decisions, progress, dependencies, blockers, next steps, exploration summaries, large textual tool results, or selective historical…

phenomenoner/Chatgpt-Codex-App-Plus · 88 tokens

aitask-note

Send a durable note to an existing aitask — context that task needs which is not itself work. Use when you learn something a task that already exists depends on, such as a stale assumption in its body, a wider blast radius, or a decision that changes its approach.

beyondeye/aitasks · 59 tokens

compound