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 ash1794/vibe-engineering --skill vibe-reflect-and-compoundgit clone --depth 1 https://github.com/ash1794/vibe-engineeringWrote 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/ash1794/vibe-engineering/vibe-reflect-and-compound)<a href="https://agentmods.dev/skills/ash1794/vibe-engineering/vibe-reflect-and-compound"><img src="https://agentmods.dev/badge/skills/ash1794/vibe-engineering/vibe-reflect-and-compound/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/ash1794/vibe-engineering/vibe-reflect-and-compound"><img src="https://agentmods.dev/badge/skills/ash1794/vibe-engineering/vibe-reflect-and-compound.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 | $0.00070 | $0.00824 |
| Opus 5.5 | $0.00028 | $0.00330 |
| Sonnet 5.5 | $0.00014 | $0.00165 |
| Haiku 4.5 | $0.00007 | $0.00082 |
Grade A, and why
vibe-reflect-and-compound 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 18d 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
vibe-reflect-and-compound
Every problem solved should leave something behind that makes the next one cheaper. This skill records it in the one place the agent will actually read next session.
When to Use This Skill
- After receiving feedback on your work (positive or negative)
- After resolving a non-trivial bug: it took more than ~10 minutes, was hard to reproduce, or had a surprising root cause
- After a failed attempt or an approach that had to be abandoned
- When you implement the same code pattern for the third time
- At the end of a complex task
When NOT to Use This Skill
- After trivial tasks (typos, renames, obvious fixes)
- When nothing new was learned
- During active implementation (reflect after, not during)
- For architectural decisions (use
vibe-decision-journal, which keeps ADRs in the repo)
Where to Save
Write to the memory file the current agent loads automatically. A separate LEARNINGS.md that nothing loads is write-only.
| Agent | Loaded automatically |
|---|---|
| Claude Code | CLAUDE.md (project) or auto-memory, if enabled |
| OpenAI Codex | AGENTS.md |
| Gemini CLI | GEMINI.md, or the built-in memory tool |
| Shared across tools | AGENTS.md, imported or referenced from the others |
Keep the always-loaded file short. Put long entries in docs/learnings/ (bugs, patterns, learnings) and add a one-line pointer to them in the memory file.
Entry Types
1. Learning
- **[Pattern name]**: [actionable rule]. Applies when: [context]. Evidence: [what happened].
Be actionable and specific. Not "be careful with concurrency" but "run the race detector before committing code that touches sessionCache."
2. Bug
### [Category]: [short description] ([date])
Symptom: [what was observed]
Root cause: [the actual problem]
Why hard to find: [what misled the investigation]
Fix: [what changed] — Prevention: [test, lint rule, or guard added]
Files: [paths]
Categories: Concurrency, State, Validation, Integration, Environment, UI.
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.
- 18d ago First seen · 86 lines · 70 tokens per session scan A a0d186bb5a29
vibe-reflect-and-compound is a skill published in the GitHub repository ash1794/vibe-engineering (164 stars, last pushed 6d ago), licensed MIT. It adds 70 tokens to every session and 824 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 2026-09-26.
Other skills, from other repositories
apply
Use when a staged proposal has been reviewed and signed off — REVIEW.md Step 3 ticked reviewed or provisional — and is ready to promote into the live contextualizer.
gut
Godot Unit Testing (GUT) framework for GDScript. Covers test structure, assertions, doubles/mocks, async testing, memory management, and CLI execution. Use when writing tests, understanding test patterns, or troubleshooting test failures.
report
Writes the session final report to a file, then prints only the path and a one-line summary. Fires when the prompt contains "Report per memstack:report", and also when the prompt begins with a standing trigger configured through MEMSTACKREPORTONTASKPROMPTS or MEMSTACKREPORTTRIGGERS. Dormant otherwise.
verify
Runs this project's check chain through scripts/verify.py and reads the receipt it writes. Fires when tracked changes are finished, when the user asks whether work passes, before a commit, and before reporting a task done. Stays dormant in repositories with no detectable check chain, during read-only audits, and for…
memstack-development-webapp-testing
Use when the user says 'write browser tests', 'test this page', 'playwright test', 'e2e test', 'end to end test', 'browser test', 'test the UI', or needs Playwright-based browser testing for a web application. Do NOT use for unit tests, API tests, or non-browser testing.
token-optimization
Use when the user says 'token optimization', 'save tokens', 'context window', 'reduce tokens', 'token stack', or 'TokenStack', or asks about extending context window capacity. Covers TokenStack, the built-in compression proxy that shrinks Claude Code tool output before it reaches the Anthropic API. Do NOT use for…