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 hamr0/liteagents --skill remembergit clone --depth 1 https://github.com/hamr0/liteagentsWrote 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/hamr0/liteagents/remember)<a href="https://agentmods.dev/skills/hamr0/liteagents/remember"><img src="https://agentmods.dev/badge/skills/hamr0/liteagents/remember/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/hamr0/liteagents/remember"><img src="https://agentmods.dev/badge/skills/hamr0/liteagents/remember.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 5 findings, up to high
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 →
- high Prompt Injection · line 66 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Memory Poisoning · line 478 Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.Fix: Protect agent memory and state from modification by untrusted content. Use read-only memory for critical instructions and validate all state changes.
- medium Excessive Agency · line 65 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Rogue Agent · line 94 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.
- medium Rogue Agent · line 292 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 | $0.00012 | $0.06470 |
| Opus 5.5 | $0.00005 | $0.02588 |
| Sonnet 5.5 | $0.00002 | $0.01294 |
| Haiku 4.5 | $0.00001 | $0.00647 |
Grade B, and why
remember scanned grade B with 1 finding 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 19d 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.
Subtle steeringmediumPrompt injection
Instructions that bias recommendations or shape behaviour without the user noticing.
- **Not set up** — no `docs/.docs-builder/` directory (docs-builder never ran), or `due` prints "no ledger yet": print one line telling the user to run `/docs-builder reorg`. Never tell them to run `ledger` — it would st Copies of this mod
1 near-identical copy found in the catalogue:
- remember — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run friction analysis, then consolidate .amp/stash/*.md + friction antigens into one project-local .amp/remember/MEMORY.md and inject it into AGENT.md. Friction runs automatically (best-effort) — there is no separate /friction command. A detect-only docs check runs at the end. Why each rule exists: docs/product/remember-README.md.
Rules
- Simple and scoped: minimal implementation, only the requested outcome.
- Scripts: call every bundled script (
friction.cjs,version-check.cjs,sync-rules.cjs,stub-check.cjs,docs-builder.cjs) by its absolute path — the cwd is the target repo, not this package, so a cwd-relative path fails. - Precision over recall: a false antigen loaded via
@.amp/remember/MEMORY.mdsteers every future session — unsure → do not promote; leave it to recurrence. - Mid-tier model: steps 2, 3 and 4a use your tool's balanced mid tier — never the cheapest tier, never a vendor model name.
- Batch stashes: up to 5 stashes per extraction agent, as few agents as possible (3 → 1 agent, 7 → 2); if more than one, run them concurrently.
- Facts are short rules, not prose: one line, target 160 chars, hard stop 180, stating a rule that changes future behaviour. Events and history are not facts.
- Never re-read session logs: work from the quotes in
antigen_clusters.json. - Never hand-compute counts:
friction.cjs countdoes them. - Relay script output verbatim, never re-worded (a paraphrase loses the filename the user needs). This covers output meant for the user; JSON a step consumes (e.g.
friction.cjs count) is input, not something to relay. - Tiers (defined once): High = 5+ distinct sessions → loaded hot via
@MEMORY.md; Medium = 3-4 → recorded under Antigens, not loaded; Low = under 3 → ledgerobservingat 2 sessions, nothing at 1. Grouping consolidates evidence, never elevates it. Never auto-promote below High.
Steps
-
Run friction first (best-effort — friction analyzes ALL your usage, so point it at the tool's global sessions root, not a per-project directory)
- Locate
friction.cjs— bundled atremember/friction.cjsbeside this file (installed or run from the package). If it exists nowhere, skip to step 1 (stash-only) and tell the user it is missing. - Check for a newer liteagents (best-effort, one line, never blocking):
If that path does not exist, use the directory you resolved fornode ~/.config/amp/skills/remember/version-check.cjsfriction.cjs(the two ship side by side). It prints one advice line if the install is behind the registry, else nothing; it exits 0 always, caches 24h, and is bounded to ~2s. If it prints a line, relay it verbatim in the final report; never act on it, never run the install yourself. - Script missing from both locations → say so (one line). A failed check (offline, registry down, timeout) stays silent by design; a missing script means an incomplete install.
- Resolve the global sessions root — probe top-to-bottom, use the first that exists and holds
.jsonlfiles directly or one level down in per-project subdirectories (friction scans exactly those two levels). Never prompt the user.# ── Add your own global sessions root at the TOP so it is checked first ── ~/.claude/projects/ # Claude Code ~/.factory/projects/ # Droid / Factory ~/.config/amp/projects/ # Amp ~/.config/opencode/projects/ # opencode ~/.codex/sessions/ # Codex CLI (use $CODEX_HOME/sessions/ if set) ~/.gemini/antigravity-cli/brain/ # AntigravityNote:
friction.cjsparses Claude Code's session schema. The Codex/Antigravity roots will resolve but yield no signals until friction learns their formats — open an issue to request one: https://github.com/hamr0/liteagents/issues - Run
node <friction.cjs> "<resolved-root>". It writes to.amp/remember/friction/in the current project. Run it as its own command — nothing chained before or after it — with a 10-minute tool timeout: it scans the whole sessions root and can take a minute or more. - On any miss — loud, never silent. If no root resolves, friction errors, or it finds no usable sessions, print this and continue stash-only:
⚠️ Friction didn't run — no sessions found. To enable it, open this skill's own file and add your tool's global sessions root to the TOP of the probe list in step 0, then re-run
/remember. Consolidating stashes only this time.
- Locate
What ships with it
5 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.
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.
- 19d ago Changed b15e412387bd
- 20d ago Changed · -367 lines scan A → B 628638ab2d3d
- 1mo ago First seen · 626 lines · 12 tokens per session scan A 83de2111a0b7
remember is a skill published in the GitHub repository hamr0/liteagents (26 stars, last pushed today), licensed Apache-2.0. It adds 12 tokens to every session and 6,470 once invoked, about $0.0000 per session on Opus 5.5. A static security scan graded it B with 1 finding (subtle steering). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.
Other skills, from other repositories
persist-memory
Persist information to long-term memory at /.clacky/memories/. Use when the user asks you to remember/note something, or when reviewing a finished conversation for facts worth keeping. Handles file naming, topic merging, frontmatter, and size limits.
recall-memory
Recall relevant long-term memories on demand. Given a topic or question, judges relevance from pre-loaded metadata, loads only relevant files, and returns a concise summary to the main agent.
doris-debug-resource-isolation
Use for Doris Workload Group / resource tag queue starvation, CPU/memory isolation leaks, and workload policy debugging. Commands: SHOW WORKLOAD GROUPS, EXPLAIN resource.
find-skills
Use when automatically discover, evaluate, and activate community skills when local skills don't cover user needs. Includes credibility scoring and safety checks for complete OpenClaw self-sufficiency.
logging-session
Record and query AI conversation logs — what users asked, how it was solved, and the result. Use when users want to logging conversation, summarize recent work from sessions, or want daily/weekly project summaries from past conversations.
personal-ip-knowledge-builder
A tool for turning interviews, chats, résumés, public content, business materials, case studies, and existing documents into a long-term personal knowledge base for an AI-assisted public profile.