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 CloseTheLoops/moradin --skill retrospectgit clone --depth 1 https://github.com/CloseTheLoops/moradinWrote 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/closetheloops/moradin/retrospect)<a href="https://agentmods.dev/skills/closetheloops/moradin/retrospect"><img src="https://agentmods.dev/badge/skills/closetheloops/moradin/retrospect.svg" alt="Measured on agentmods" 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 | $0.00025 | $0.00560 |
| Opus 5 | $0.00013 | $0.00280 |
| Sonnet 5 | $0.00005 | $0.00112 |
| Haiku 4.5 | $0.00003 | $0.00056 |
Grade A, and why
moradin-retrospect 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 6d 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
moradin:retrospect
Review past work across a time window. Identify what's worth promoting.
When invoked
/moradin:retrospect [<project>] [<window>]
Examples:
/moradin:retrospect godtech 2w— review GodTech sessions in last 2 weeks/moradin:retrospect— review all projects, all time
What you do
-
Gather session logs. Read
projects/<project>/sessions/*.md(or all projects). Filter by date if window specified. -
Read recent lessons under
memory/lessons/. Especially ones cited multiple times in sessions. -
Identify recurring patterns:
- Did the same lesson get applied 3+ times? → candidate for promotion to a pattern.
- Did the same pattern get applied across multiple projects? → candidate for promotion to a universal principle.
- Did the same kind of mistake keep happening? → candidate for a new principle.
-
Propose promotions to operator:
- "Lesson
check_code_first.mdwas cited in 4 sessions. Promote to principle?" - "Pattern
harness_contract.mdapplied to GodTech 3x and proposed for trading. Already widely used — consider locking the spec." - "I notice you've hit the same Verifier-bypass issue 3 times. Make it a principle?"
- "Lesson
-
For each approved promotion:
- If lesson → pattern: copy to
memory/patterns/, update frontmatter (type=pattern), keep original lesson file as reference - If pattern → principle: copy to
memory/principles/, update frontmatter (type=principle), keep pattern
- If lesson → pattern: copy to
-
Identify decay candidates:
- Patterns not referenced in any session for 90+ days
- References whose source URL changed materially (run
scripts/audit_references.py)
-
Update indexes and produce a retrospective summary.
Don't
- Don't auto-promote without operator approval.
- Don't delete files during retrospective. Mark with
superseded_by:if a newer version exists. - Don't propose promotion based on a single occurrence. Wait for the 3rd.
Output
- Retrospective summary (counts, recurring themes, promotion candidates)
- 0-N new files in
memory/patterns/ormemory/principles/(after operator approval) - 0-N files marked
superseded_by:in their frontmatter
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.
- 6d ago First seen · 55 lines · 25 tokens per session scan A 8c9b4f8afbcf
moradin-retrospect is a skill published in the GitHub repository CloseTheLoops/moradin (2 stars, last pushed 28d ago), licensed MIT. It adds 25 tokens to every session and 560 once invoked, about $0.0001 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
context-manager
Use when a long-running agent task needs context budgeting, checkpointing, compaction, retrieval, or capability-based model routing.
knowledge-ops
Use when an Obsidian knowledge system needs classification, deduplication, retrieval, synchronization, debt queues, or governed Agent/Wiki promotion.
session-learn
Use when a completed work session should yield durable concepts, corrections, decisions, reusable patterns, and a traceable next action.
log-session
Append a structured entry to the project's session log (docs/LOGS.md): what was done this session, files touched, decisions taken, and the next step — so the next session (or another person) can pick up the thread without re-reading git history. Run it before /clear, before closing Claude Code, or at any natural…
context-engineering
Workflow for context packets, context audits, compaction, handoffs, session memory, and deciding what AI-agent context to load, retrieve, trim, summarize, refresh, or omit. Use for context rot, context flooding, stale or missing context, task switching, and long-running agent sessions. Do not use when prompt wording…
wiki-lint
Health-check a wiki vault. Finds orphan pages (no inbound links), dead wikilinks (point to non-existent pages), missing frontmatter fields, stale claims, empty sections, and pages absent from catalog.md. Produces a structured report with severity tiers and proposes concrete fixes — but does not auto-apply them unless…