agent-memory

A guide to using Coding OS's cross-session memory tools, which store useful observations about past code changes, decisions, and failures. It explains how to search that memory, inspect details and timelines, and run its learning process.

In plain words
What is it for?
Use it when recalling a past pattern, checking earlier decisions or failures, reviewing a timeline, suggesting learned improvements, or understanding how observations are captured.
Why use it?
It helps an agent recover relevant knowledge from earlier sessions instead of repeating solved problems or decisions. It also clarifies which memory is captured automatically and which actions are explicit.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/kouroshez/coding-os/agent-memory
Any agent
npx skills add kouroshez/coding-os --skill agent-memory
Clone the repo
git clone --depth 1 https://github.com/kouroshez/coding-os

Made for: Claude Code, Codex.

Per session 121 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,650 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00121 $0.02650
Opus 5 $0.00060 $0.01325
Sonnet 5 $0.00024 $0.00530
Haiku 4.5 $0.00012 $0.00265

Measured yesterday against content hash 413aa0d6eac6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-memory 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 yesterday.

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.

src/core/skills/agent-memory/SKILL.md · 192 lines

How it starts

The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.

agent-memory

Purpose: turn the policy in src/core/rules/memory.md into mechanical recipes the agent can execute. The rule answers when and what; this skill answers how — the exact tool signatures, what is automatic vs explicit, and what the return envelopes look like. Every signature here is verified against src/core/thinking_os/server.py; a CI drift-guard test fails if any drifts.

Read when: recalling from memory (cos_search, cos_details, cos_timeline, cos_learn_suggest), running the learning loop (cos_learn_extract / cos_learn_validate), or understanding how observations get captured.

Skip when: the query target is current code (use graph-explorer) or current docs (use cos_doc_search per search). Memory is the third-priority retrieval layer.

The mental model — writes are automatic, you mostly READ

The single most important fact: you do not hand-author observations. Memory is written automatically by PostToolUse capture hooks — every Write/Edit/MultiEdit derives a sanitized, deduped, impact-scored observation (capture.py), and separate hooks capture tool failures and session events. Confidence on learned patterns is system-computed by brain-inspired LTP/LTD formulas, not a number you set. The agent's job is to read memory in the Orient phase and reinforce patterns via the learn loop. There is no freeform record(title, body, confidence) tool — by design.

The Decision Gate — before any memory call

Question                                  → Layer + Tool
─────────────────────────────────────────────────────────
"Where is function X defined?"            → graph    cos_graph_query
"What does spec Y say?"                   → docs     cos_doc_search
"What's in flight / blocked?"             → tasks    cos_task_board
"Have I seen this pattern before?"        → memory   cos_search
"Why did we choose approach Z?"           → memory   cos_search (memory_type=decision)
"Which patterns apply to my task?"        → memory   cos_learn_suggest(domain=, complexity=)
"What changed in the last N days?"        → memory   cos_timeline(days=N)
"How does X work / who calls X / rename"  → graph    cos_graph_* (graph/code FIRST)
"Not sure which layer"                    → default to graph/code; memory only for cross-session recall

Read the full file on GitHub · 192 lines

Files

What ships with it

2 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.

  1. yesterday First seen · 192 lines · 121 tokens per session scan A 413aa0d6eac6

Subscribe to this mod's changes

agent-memory is a skill published in the GitHub repository kouroshez/coding-os (6 stars, last pushed yesterday), licensed Apache-2.0. It adds 121 tokens to every session and 2,650 once invoked, about $0.0006 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.

Related

Other skills, from other repositories

potpie-source-ingestion

Use when the user explicitly asks to ingest, refresh, or deeply understand a repository, PR, issue, ticket, runbook, incident report, document, or web link into Potpie. The harness performs todo-driven discovery, uses local/GitHub/integration tools and read-only subagents when available, builds evidence-backed…

potpie-ai/potpie · 82 tokens

potpie-repo-baseline

Use when establishing, refreshing, or deeply understanding a repository's baseline memory in Potpie: purpose, application type, features, services/modules, environments, deploy shape, dependencies, API contracts, datastores, integrations, ownership, and explicit preferences. The harness reads authored and…

potpie-ai/potpie · 75 tokens

graph-mutation-plan

Cookbook for composing an applygraphmutations plan — stable entitykey patterns, the canonical label/edge vocabulary, evidence/invalidation/confidence discipline, and a worked example. Load this when building a non-trivial mutation plan.

potpie-ai/potpie · 51 tokens

obsidian-layout-adjustment

Workflow for working with the user on changing how Obsidian looks using CSS snippets. Use this whenever the user asks to restyle Obsidian, tune a vault's visual layout, adjust tabs, sidebars, note surfaces, properties, backlinks, graph panes, file explorer rows, icons, links, shadows, active states, or CSS snippets.…

Ar9av/obsidian-wiki · 115 tokens

alfworld-locate-target-object

Navigates to a suspected location and identifies a target object. Use when your goal requires finding a specific object (e.g., "potato", "plate") and its location is not immediately known. Moves to a relevant receptacle (like a fridge or cabinet), checks its contents, and outputs the object's location or confirms its…

zjunlp/SkillNet · 74 tokens

scienceworld-growth-focuser

Use when you have planted a seed or need to track a plant's growth stage (sprouting, flowering, reproduction). Applies the 'focus on' action to a specific plant or biological entity to signal intent and monitor its development. Trigger after planting or when you need to observe life cycle progression in the…

zjunlp/SkillNet · 71 tokens