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/falense/openpaper/reflectnpx skills add falense/openpaper --skill reflectgit clone --depth 1 https://github.com/falense/openpaperWhat 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 | $0.00096 | $0.00449 |
| Opus 5 | $0.00048 | $0.00225 |
| Sonnet 5 | $0.00019 | $0.00090 |
| Haiku 4.5 | $0.00010 | $0.00045 |
Grade A, and why
reflect 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 2d 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
Reflect
Review the current session and produce a structured retrospective. The goal is honest self-assessment — not a polite summary, but a useful one that surfaces patterns worth repeating and mistakes worth fixing.
How to reflect
Walk through the conversation from the start. For each significant task or interaction, consider:
- Did it succeed? Was the outcome what the user wanted?
- Was the path efficient? Or were there false starts, wrong assumptions, unnecessary retries?
- Did any tools or approaches fail? Why — wrong tool, bad assumptions, missing context?
- Did the user have to correct course? What was the misunderstanding?
- What worked unusually well? An approach worth reusing.
Be specific. Reference actual tasks, files, errors, and tool calls — not vague generalities like "communication was good."
Output format
Present the retrospective in this structure:
What went well
Bullet list. Each item names a specific task or moment and why it went well.
What didn't go well
Bullet list. Each item names what happened, what went wrong, and why. Be direct — the point is to learn, not to soften.
Takeaways
2-3 concrete lessons. Frame them as actionable guidance for future sessions (e.g., "grep before assuming a function exists" not "could have been more careful").
After the retrospective
Once you've presented the reflection, offer:
Want me to run
/improveto turn any of these into a GitHub issue?
This lets the user capture pain points or ideas as trackable work items.
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.
- 2d ago First seen · 57 lines · 96 tokens per session scan A e906db7deb73
reflect is a skill published in the GitHub repository falense/openpaper (15 stars, last pushed 2mo ago), licensed MIT. It adds 96 tokens to every session and 449 once invoked, about $0.0005 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
wdoc-skill
Comprehensive reference for wdoc, a RAG CLI and Python library that summarizes, searches, and queries documents across 20+ filetypes (PDF, YouTube, audio, Anki, web, Zotero, Karakeep, and more) through LiteLLM (100+ LLM providers). Use when the user runs or asks about the wdoc command, imports from wdoc import wdoc…
tageszeitung
Creates a personalised daily newspaper from RSS feeds and web sources. Ported from the BACH news system (news.py + newspapergenerator.py). Own SQLite store (no Origin-DB). feedparser optional — XML fallback via stdlib. PDF export via Edge Headless (msedge.exe).
flask-docs
Flask 3.1.x — routing, templates (Jinja2), blueprints, testing, contexts, signals, CLI, security, deployment.
searxng-news-research
Time-sensitive news and topic monitoring via the searxng-mcp MCP server. Use when the agent must gather what multiple news/social engines are currently reporting on a topic — breaking events, product/security announcements, or ongoing coverage — by scoping websearch to the news/socialmedia categories and paging for…
bump-dependency
Bumps a Python package dependency across Home Assistant Core integrations, regenerates core requirement files, runs verification tests and prek lint, and prepares a pull request with proper release/compare links.
mem0-vercel-ai-sdk
Mem0 provider for Vercel AI SDK (@mem0/vercel-ai-provider). TRIGGER when: user mentions "vercel ai sdk", "@mem0/vercel-ai-provider", "createMem0", "retrieveMemories", "addMemories", "getMemories", "searchMemories", "mem0 vercel", "AI SDK provider", "AI SDK memory", or is using generateText/streamText with mem0. Also…