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 ahmedtawfeeq1/loredex --skill loredex-curategit clone --depth 1 https://github.com/ahmedtawfeeq1/loredexWrote 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/ahmedtawfeeq1/loredex/loredex-curate)<a href="https://agentmods.dev/skills/ahmedtawfeeq1/loredex/loredex-curate"><img src="https://agentmods.dev/badge/skills/ahmedtawfeeq1/loredex/loredex-curate/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/ahmedtawfeeq1/loredex/loredex-curate"><img src="https://agentmods.dev/badge/skills/ahmedtawfeeq1/loredex/loredex-curate.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 | $0.00074 | $0.00497 |
| Opus 5.5 | $0.00030 | $0.00199 |
| Sonnet 5.5 | $0.00015 | $0.00099 |
| Haiku 4.5 | $0.00007 | $0.00050 |
Grade A, and why
loredex-curate 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.
What it actually says
/loredex-curate
Ask two questions before running anything:
- Objective — "What's the objective this brief should answer?" (e.g. "draft the spec
for the P0 agent"). The brief, reading order, and next actions are all steered by it.
No objective given → omit
--objectiveand the LLM derives one from the notes. - Scope — the whole project, or a recent task batch? A session that just produced N
files is task scope:
--since <date>or its--topic(s). Scoped runs write a separateStart Here - <project> - <slug>.md(a durable session handoff); unscoped runs overwrite the mainStart Here - <project>.md.
Then:
npx -y loredex@latest curate <project> --objective "<text>" [--since <date>] [--topic <t>] --dry-run
Summarize the printed plan out loud — objective, reading order, stale flags, merge
candidates, orphan/drift counts — then re-run with -y once they confirm.
Every run also does this automatically, no extra command, with or without an LLM:
- Ghost-link cleanup:
[[x.py]]-style links to code get rewritten so Obsidian's graph doesn't fill with phantom nodes. - Drift detection: notes whose
source_pathfile has git commits after the note's owndateget auto-stampedstale. - Orphan detection: notes nobody links to are reported (not auto-fixed) — worth mentioning to the user.
If the project has grown large and this feels slow, mention --max-detailed <n> (default
60) — only that many of the most recent notes get full content in the prompt; older ones
become a metadata-only index instead.
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 · 38 lines · 74 tokens per session scan A c6b22c476926
loredex-curate is a skill published in the GitHub repository ahmedtawfeeq1/loredex (2 stars, last pushed 3mo ago), licensed MIT. It adds 74 tokens to every session and 497 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
karpathy-llm-wiki
Use when building or maintaining a personal LLM-powered knowledge base. Triggers: ingesting sources into a wiki, querying wiki knowledge, linting wiki quality, 'add to wiki', 'what do I know about', or any mention of 'LLM wiki' or 'Karpathy wiki'.
knowit
Persistent, structured project memory for AI coding agents via the Knowit MCP server. Use before planning or editing code to load relevant rules, architecture, decisions, patterns, and conventions; use after finishing work to save durable learnings. Triggers on "remember this", "what are our conventions", "project…
shared-agents-knowledge
Sync and use the shared-agents repo (team skills, rules + learnings). Use at session start, when loading team knowledge, searching learnings, or when the user mentions shared-agents, shared learnings, or team knowledge.
unforgit-memory
Use Unforgit MCP tools as durable repository memory for decisions, conventions, gotchas, and playbooks.
debrief
Use at the end of a session (or midway through a long one) to deliberately capture what was learned — lessons, decisions, workflows, state changes, and facts — into the knowledge base. The nightly harvest is a safety net for lessons; unless the host opted into KBHARVESTFACTS, /debrief is the only thing that records…
kb-workflow
Use before starting any non-trivial task to search the knowledge base for context, and after completing work to capture findings. Teaches the token-efficient retrieval pattern and self-learning loop.