memory-loop

A task workflow that uses Aionforge Memory, an external service for saving and retrieving project context, decisions, and work items.

In plain words
What is it for?
Implementation, debugging, code review, releases, incident response, planning, handoffs, and other multi-step work where durable context matters.
Why use it?
It helps coding sessions continue from earlier decisions and unfinished work instead of starting with no context after each handoff or interruption.

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/jscott3201/aionforge-memory/memory-loop
Any agent
npx skills add jscott3201/aionforge-memory --skill memory-loop
Clone the repo
git clone --depth 1 https://github.com/jscott3201/aionforge-memory

Made for: Claude Code, Codex.

Per session 47 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 704 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 $0.00047 $0.00704
Opus 5 $0.00023 $0.00352
Sonnet 5 $0.00009 $0.00141
Haiku 4.5 $0.00005 $0.00070

Measured 2d ago against content hash f799a430fce1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

memory-loop 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 2d 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.

plugins/aionforge-memory/skills/memory-loop/SKILL.md · 37 lines

How it starts

The opening of the file, as written. The whole thing — 37 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Memory Loop

Requires an enabled Aionforge Memory MCP server.

Use this skill to make memory part of the task loop, not a final afterthought.

Procedure

  1. Start with memory-recall. Search broadly enough to find prior decisions, preferences, blockers, release state, and failed attempts. Also recall open work with work_query / work_tree so you continue the backlog instead of re-deriving it. On every read path — search, read_memory, session_manifest, work_query — assert the teams you belong to (e.g. teams: ["aionforge-memory-team"]); read authorization is per-call, so team memory and team work items are out of scope unless you assert the team on that call. A by-id read_memory of a team memory needs that team asserted in the same call (parity with search) — it never auto-widens.
  2. Work from current evidence. Recalled memory can guide attention, but repo state, tool output, and user instructions win.
  3. Capture along the way the moment a durable fact lands: decision made, blocker found, fix verified, release changed, user preference learned, or approach rejected. Do not save these for the end — a context compaction can discard them first.
  4. Track the work as it moves. When a task, blocker, or TODO appears, work_create a work item (see the work-tracking skill); work_advance its status as it progresses. Tasks are work items, not memory episodes — and there is no "note" to store directly.
  5. Be generous with memory. Aionforge can handle large memory sets; several precise records are better than one vague end note.
  6. At natural checkpoints, search again if new terms, file paths, ids, or failures appear.
  7. Before ending, capture a handoff when future agents would benefit: branch, PR, commits, tests, CI, remaining work, and caveats. Leave the remaining work as work items so the next agent can work_query it. When the handoff is directed at a specific agent or teammate — not just left for whoever comes next — also message_send them a brief so they actually receive it (see the agent-messaging skill): the capture and work items make it durable, the message delivers it.
  8. Run consolidation_status; run consolidate only when the approval policy permits mutating derived memory.

Read the full file on GitHub · 37 lines

Files

What ships with it

1 file 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. 2d ago First seen · 37 lines · 47 tokens per session scan A f799a430fce1

Subscribe to this mod's changes

memory-loop is a skill published in the GitHub repository jscott3201/aionforge-memory (10 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 47 tokens to every session and 704 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.

Related

Other skills, from other repositories

zettelforge

ZettelForge v2.0.0 — Production CTI agentic memory system. Hybrid TypeDB (STIX 2.1 ontology) + LanceDB (vector search). Zero external AI dependencies: fastembed for embeddings, llama-cpp-python for LLM. 75% accuracy on CTI queries, 18% on LOCOMO. Use when agents need persistent memory, threat intel retrieval, entity…

ThreatRecall/zettelforge · 97 tokens

matryca-github

Apply Matryca maintainer standards for GitHub issues, pull requests, branches, reviews, merges, milestones, tags, releases, and remote comments. Use before any GitHub-facing action or artifact.

MarcoPorcellato/matryca-plumber · 47 tokens

matryca-changelog

Decide whether a completed Matryca change belongs in CHANGELOG.md and add one concise Unreleased entry when required. Use before concluding runtime, security, architecture, integration, performance, operator, or public-contract changes.

MarcoPorcellato/matryca-plumber · 50 tokens

matryca-clean-architecture

Preserve Matryca Clean Architecture dependency direction, thin surfaces, domain ownership, typed boundaries, config parsing, and scoped refactors. Use for module moves, shared abstractions, layer imports, configuration architecture, or structural Python changes.

MarcoPorcellato/matryca-plumber · 54 tokens

matryca-logseq-paradigm

Apply Matryca's Logseq OG outliner, block/property, namespace, and Markdown semantics. Use for graph-content generation, Logseq block or page-property changes, namespace handling, or code that transforms vault content.

MarcoPorcellato/matryca-plumber · 53 tokens

matryca-prompt-maintainer

Maintain Matryca prompt builders, Tier-2 OpenSpec fragments, generated SYSTEMPROMPT.md, prompt hashes, MCP tool docstrings, and llms.txt synchronization. Use for any prompt or MCP tool-contract change.

MarcoPorcellato/matryca-plumber · 52 tokens