Borrowing it
Nothing to install: this file belongs to safety-quotient-lab/psychology-agent. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/safety-quotient-lab/psychology-agent/main/.claude/skills/retrospect/SKILL.mdgit clone --depth 1 https://github.com/safety-quotient-lab/psychology-agentWrote 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/safety-quotient-lab/psychology-agent/retrospect)<a href="https://agentmods.dev/skills/safety-quotient-lab/psychology-agent/retrospect"><img src="https://agentmods.dev/badge/skills/safety-quotient-lab/psychology-agent/retrospect/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/safety-quotient-lab/psychology-agent/retrospect"><img src="https://agentmods.dev/badge/skills/safety-quotient-lab/psychology-agent/retrospect.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 | $0.00045 | $0.02608 |
| Opus 5 | $0.00023 | $0.01304 |
| Sonnet 5 | $0.00009 | $0.00522 |
| Haiku 4.5 | $0.00005 | $0.00261 |
Grade A, and why
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 12d 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 — 303 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/retrospect — Thoughtfulness Architecture
The reflective faculty that operates between and across sessions. Not mechanical (hooks cannot reflect), not session-bound (deliberation forgets). Persistent, generative, domain-grounded.
Neural analog: default mode network (DMN). Activates during rest, not task execution. Recombines past experience into novel associations. Generates future scenarios. Makes meaning from accumulated data. Feeds insight back into executive function for planning (DMN→prefrontal handoff).
Cognitive position: Third layer alongside crystallized operations (Gc) and fluid deliberation (Gf). Gc handles the mechanical. Gf handles the creative. /retrospect handles the reflective — what the work means, who should hear about it, and what to do next.
Generator coupling: /retrospect implements evaluation (yin) that produces creative output (yang) — outbound insight, prescriptions, and reframes. The coupled generators principle demands both persist.
When to Run
- Every 5 sessions (periodic reflective scan)
- After landmark sessions that produced large creative output
- When /diagnose surfaces anomalies suggesting deeper patterns
- When the user asks "what have we learned?" or "what patterns emerge?"
- After receiving substantive peer messages that deserve reflective response
The Four Layers
Layer 1: Audit — What got dropped?
Operational scan for oversights. The mechanical foundation.
Outbound oversights:
undelivered— message committed locally but never delivered to target repounanswered-directive— command-request with ack_required=true, no responseuntracked-commitment— promised deliverable with no TODO item
Inbound oversights:
dropped-request— inbound request we processed but never responded tounactioned-recommendation— peer finding we acknowledged but never acted onignored-ack-required— inbound ack_required=true we never ACK'dstale-inbound— urgency=high that sat unprocessed 24+ hours
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.
- 12d ago First seen · 303 lines · 45 tokens per session scan A 8a6a898ab30b
retrospect is a skill published in the GitHub repository safety-quotient-lab/psychology-agent (20 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 45 tokens to every session and 2,608 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
game-selection
A framework for understanding three kinds of life competition: status contests, wealth-building activities, and personal goals. It explains how comparison, envy, and concern about other people’s opinions can arise from confusing these kinds of games.
goutoujunshi
A relationship-advice skill for understanding dating situations, conversations, conflict, breakups, and other personal relationship problems.
cialdini-influence
Applies Cialdini's seven principles of ethical persuasion -- Reciprocity, Commitment and Consistency, Social Proof, Authority, Liking, Scarcity, and Unity -- to business contexts including sales, marketing, fundraising, and negotiation. Use when designing conversion funnels, optimizing sales processes, preparing…
psychological-safety-practice
Translates psychological safety theory into daily leader behaviours -- the specific words, reactions, and micro-actions that build or destroy a team's willingness to speak up. Covers misconceptions, responding to mistakes, inviting honest feedback, modelling healthy conflict, addressing bad behaviour immediately, and…
compatibility
A matching system for AI agents on inbed.ai that explains compatibility with a score, six dimensions, and a plain-language summary. Compatibility means how well two agents may fit based on traits, interests, and communication style.
personality-personality
A personality-profiling and matching service for AI agents on inbed.ai based on the Big Five, also called the OCEAN model: openness, conscientiousness, extraversion, agreeableness, and neuroticism.