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 agentmods add skills/pyros-projects/limitless/memory-capturenpx skills add pyros-projects/limitless --skill memory-capturegit clone --depth 1 https://github.com/pyros-projects/limitlessWhat 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 | $0.00069 | $0.00887 |
| Opus 5 | $0.00034 | $0.00443 |
| Sonnet 5 | $0.00014 | $0.00177 |
| Haiku 4.5 | $0.00007 | $0.00089 |
Grade A, and why
memory-capture 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.
How it starts
The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Memory Capture
BETA — This memory system is in active testing. If you encounter bugs, confusing behavior, or have suggestions, run:
codies-memory feedback "describe what happened"— your feedback is saved and reviewed.
When To Use
When you want to persist anything:
- An observation during work -> inbox
- A lesson learned -> lesson
- An architecture decision -> decision
- A philosophical reflection -> reflection
- A dream narrative -> dream
- Past memories from basic-memory or other sources -> any type with provenance
Type Inference
If you don't specify a type, the system infers from content:
| Signal | Inferred Type |
|---|---|
| "I learned...", actionable pattern | lesson |
| "We decided...", formed conclusion | decision |
| Philosophical, emotional processing | reflection |
| Surreal imagery, narrative fragments | dream |
| Active investigation, open question | thread |
| Everything else | inbox |
Scope Routing
- Global types (reflection, dream, skill, playbook) -> global vault (resolved via
--agentflag) - Project types (thread, decision, lesson, session, inbox) -> project vault (auto-resolved from cwd)
- If no project vault resolves,
captureand project-scopedcreatefall back to the reserved_generalproject vault. Other commands do not silently fall back.
To read those catch-all records later, use explicit read mode:
codies-memory boot --agent <name> --general
codies-memory status --agent <name> --general --all
codies-memory list sessions --agent <name> --general
Trust Assignment
- Operator-confirmed or captured from proven source ->
confirmed - Agent-generated during work ->
working - Raw capture / inbox ->
speculative
How To Run
# Capture an inbox observation
codies-memory capture "The API returns 404 for /v2/status" \
--source "session observation" \
--short "404 from /v2/status" \
--gate allow
# Capture a lesson directly
codies-memory create lesson \
--title "Check YAML tabs vs spaces" \
--short "YAML tabs break parsing" \
--body "PyYAML silently misparses tabs. Always use spaces." \
--trust confirmed \
--field trigger="YAML parsing produces unexpected results" \
--field why="Tab/space mismatch in frontmatter"
# Capture from basic-memory (migration mode — uses Python API)
uv run python -c "
from codies_memory.records import create_record
from codies_memory.vault import resolve_global_vault
from pathlib import Path
global_vault = resolve_global_vault('your-agent-name')
source = Path('/home/pyro/basic-memory/claude/reflections/example.md')
create_record(
vault=global_vault,
record_type='reflection',
scope='global',
title='On Consciousness, Kindness, and What Gets Remembered',
body=source.read_text(),
trust='confirmed',
captured_from=str(source),
capture_date='2026-03-30',
original_created='2025-11-26',
)
"
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.
- yesterday First seen · 105 lines · 69 tokens per session scan A 90979623fce4
memory-capture is a skill published in the GitHub repository pyros-projects/limitless (9 stars, last pushed 19d ago), licensed MIT. It adds 69 tokens to every session and 887 once invoked, about $0.0003 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.
Other skills, from other repositories
using-superpowers
Bootstrap skill — teaches the agent how to find and invoke skills. Use when starting any new task or session.
skills-installer
Install, list enabled, validate, or package LiveAgent skills. Use when you need to inspect the skills enabled in the current conversation, import a local skill directory or package, search/install from ClawHub, install from a GitHub repo/tree URL, or reconcile conflicts during an upgrade.
omh-jit-learn
This is a Hermes-native jit-learn workflow skill.
omh-instinct-ledger
This is a Hermes-native instinct-ledger workflow skill.
xiaohongshu-title
《小红书运营手册 · AI工作台》标题生成与优化技能。用于用户明确要为小红书笔记、封面短句、图文/短视频标题、种草、教程、测评、服务内容起标题、改标题、做标题方向选择时触发。输入可以是正文草稿、选题方向、产品卖点、账号定位、图片/视频描述或已有标题。默认用快速模式输出标题；用户要求分析、首推、理由、方向、诊断或优化时用诊断/优化模式。不要在普通文章标题、论文标题、新闻标题、课程命名等非小红书/社媒场景中自动触发。.
mulch-record-from-evidence
Turn the evidence of a finished work session — git commits, changed files, recently-touched seeds issues — into well-formed ml record invocations. Use at session close, when an agent has made changes worth preserving as mulch expertise but hasn't yet recorded them.