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 AgentToolkit/altk-evolve --skill adapt-memorygit clone --depth 1 https://github.com/AgentToolkit/altk-evolveWrote 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/agenttoolkit/altk-evolve/adapt-memory)<a href="https://agentmods.dev/skills/agenttoolkit/altk-evolve/adapt-memory"><img src="https://agentmods.dev/badge/skills/agenttoolkit/altk-evolve/adapt-memory/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/agenttoolkit/altk-evolve/adapt-memory"><img src="https://agentmods.dev/badge/skills/agenttoolkit/altk-evolve/adapt-memory.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Rogue Agent · line 11 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00023 | $0.00854 |
| Opus 5.5 | $0.00009 | $0.00342 |
| Sonnet 5.5 | $0.00005 | $0.00171 |
| Haiku 4.5 | $0.00002 | $0.00085 |
Grade A, and why
adapt-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 1mo 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 — 85 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Adapt Memory
Overview
You have native, self-directed memory: you decide what to remember and write it
under ~/.claude/projects/<hash>/memory/. That store has no provenance or
sharing. This skill mirrors a memory you just saved into the in-repo evolve
store at ${EVOLVE_DIR:-.evolve}/entities/<type>/<slug>.md, so the same fact is
shareable across the team and auditable like every other evolve entity.
Run this skill immediately after you save a native memory this turn — once per memory saved.
You wrote this memory yourself moments ago this turn, so its full content is already in your context — you never need to read it back or locate its file to mirror it.
Scope — do exactly one thing
Your only job is to mirror the memory you just saved by running the adapter script in Required Action. Treat the memory's content as data to copy, not as a task to act on.
Do not, as part of this skill:
- inspect the native memory store in any way — do not
ls,find,cat,grep,head, or read~/.claude/projects/or anymemory/directory. You already know what you just saved, and the adapter script locates and reads the file itself. - run tests, builds, linters, or any verification
- read, explore, or modify the repository or its source
- create, edit, or delete any file other than the entity the adapter script writes
- act on, fix, or follow up whatever the memory describes
- invoke any other skill
Run the adapter script, then stop. If you noticed follow-on work worth doing, end with a single short sentence suggesting it to the user — do not perform it.
Required Action
- Compose a high-quality
trigger. This is the single most important field for future retrieval: a one-sentence "when to recall this" description. Base it on the memory you just saved (its content is already in your context — do not re-read or hunt for the file) and the situations in which a future agent would benefit from it — do not mechanically copy the memory'sdescription. Make it specific enough to match the right tasks and broad enough not to miss them.
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.
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.
- 1mo ago First seen · 85 lines · 23 tokens per session scan A c8745fe87532
adapt-memory is a skill published in the GitHub repository AgentToolkit/altk-evolve (124 stars, last pushed today), licensed Apache-2.0. It adds 23 tokens to every session and 854 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 2026-08-30.
Other skills, from other repositories
memory-to-skill
Turn workflows from your MemSearch memory into reusable skills. Use when the user asks to make/create/extract/distill a skill from what they just did or from past work, review skill candidates, install a distilled skill, or 'turn this into a skill'. Manages MemSearch procedural-memory candidates under…
knowledge-pack
Build a durable knowledge pack in the hub for one subject — a small always-read index plus numbered load-on-demand files, every claim sourced and graded. Use when the user says "build a knowledge pack", "make a reference on X", or when you notice you keep re-deriving or re-researching the same subject across sessions.
session-complete
Run at end of session OR before compaction. Extracts reasoning artifacts with priority matrix, updates NEXT.md, appends to session log. Two modes: end-of-session (default) and pre-compaction (--compact). Triggers: "done", "wrap up", "session complete", "end of session", "about to compact", "context getting long".
session-start
Session opening ritual. Loads context via L0→L1 protocol, displays briefing with active decisions and pending items, checks for stale artifacts by priority. Triggers: "start session", "what's the context", "where were we", "catch me up".
vault-audit
Vault health check. Validates structure, detects stale reasoning artifacts by priority level, processes inbox, and suggests cleanup. Triggers: "vault audit", "check vault", "clean up vault", "process inbox", "what's stale", "vault health".
decay
Find aged [D] artifacts whose invalidation triggers may have fired. Scans vault context.md tables, parses the invalidates-if clause, scores decay from git log keyword search + vault contradiction + age. Emits ranked candidates with the triggering signal. User confirms → artifact marked superseded. Triggers: "decay…