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 jmagly/aiwg --skill harness-memory-maintenancegit clone --depth 1 https://github.com/jmagly/aiwgWrote 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/jmagly/aiwg/harness-memory-maintenance)<a href="https://agentmods.dev/skills/jmagly/aiwg/harness-memory-maintenance"><img src="https://agentmods.dev/badge/skills/jmagly/aiwg/harness-memory-maintenance/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.
<a href="https://agentmods.dev/skills/jmagly/aiwg/harness-memory-maintenance"><img src="https://agentmods.dev/badge/skills/jmagly/aiwg/harness-memory-maintenance.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00026 | $0.01401 |
| Opus 5.5 | $0.00010 | $0.00560 |
| Sonnet 5.5 | $0.00005 | $0.00280 |
| Haiku 4.5 | $0.00003 | $0.00140 |
Grade A, and why
harness-memory-maintenance 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 today.
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 — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harness Memory Maintenance
Discover
Use aiwg memory for harness memory maintenance on every provider. Fetch only
the needed section with `aiwg show skill harness-memory-maintenance --section
Audit
Run the requested read-only inventory:
aiwg memory audit --scope project --json
aiwg memory audit --scope global --json
aiwg memory audit --scope provider --provider codex --json
Explain ownership, scope, loading chains, inaccessible or opaque surfaces and
uncertainty. Static eligibility does not prove live session loading. Use
aiwg memory cost --scope project --json for cost attribution, or
aiwg memory recommend --scope project --json for advisory candidates.
Cleanup
Audit the requested scope, then run aiwg memory recommend --scope project --json (or --scope global). Review each candidate's uncertainty, protection,
retrieval impact and owner. Cleanup selects explicit operations; it never
bulk-deletes memory. Fetch the matching operation section and Review and Apply.
Blocked candidates need owner resolution. Global recommendations are advisory;
the executable lifecycle currently supports project-owned records only.
Review and Apply
Resolve the exact owner handle from the inventory and owner records. Executable
owners are line-memory, compound-memory and semantic-memory. Show global
versus project targeting, the before/after diff, owner review requirements and
native/next-session limitations before mutation. Never reinterpret a global
request as permission to change a project or provider-native source.
Create a reviewed plan, selecting the operation's flags from its section:
aiwg memory plan --owner <owner> --target <handle> --op <operation> --rationale <reason> --reviewer <reviewer> --out <new-plan.json> --json
aiwg memory apply --plan <new-plan.json> --preview --json
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.
- today First seen · 144 lines · 26 tokens per session scan A 9ad50dbb1a94
harness-memory-maintenance is a skill published in the GitHub repository jmagly/aiwg (225 stars, last pushed today), licensed MIT. It adds 26 tokens to every session and 1,401 once invoked, about $0.0001 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 2026-10-26.
Other skills, from other repositories
hive-mind-advanced
Queen-led collective-intelligence coordination with consensus mechanisms and persistent shared memory. Use when orchestrating a large swarm of agents under a lead coordinator, building consensus across agents, or maintaining durable cross-agent memory.
lore
Curating cross-agent knowledge and institutional memory: extracts patterns from agent journals into METAPATTERNS.md, detects knowledge decay, propagates best practices. Use for memory curation.
memory-sanitizer
Scores and filters retrieved memory against trust and poisoning risk. Use for: sanitize memory, trust score retrieval, clean knowledge graph, semantic-memory-manager.
multi-platform-messenger-bridge
Unify WhatsApp, Telegram, and future channels with shared conversation context and message-quality assessment. Use for: bridge messengers, unified chat memory, cross-platform reply, WhatsApp responder.
mega-context-manager
Manages large context windows via chunking, summarization, and retrieval. Use for: manage context, too much context, chunk document, context budget.
persistent-memory-bridge
Bridges session memory to persistent storage across conversations. Use for: remember this, persistent memory, save to memory, recall later.