Agent skill

Observability

by YougLin-dev in YougLin-dev/Aha-Loop

Logs AI thoughts and decisions for human observability. An agent skill from YougLin-dev/Aha-Loop.

MITAuto-check passedDevOps & Cloud

Install Observability

skills CLI
$ npx skills add YougLin-dev/Aha-Loop --skill observability -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install YougLin-dev/Aha-Loop observability --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/YougLin-dev/Aha-Loop.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/observability .claude/skills/observability && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
observability
GitHub stars
181
Token cost
~1.7k tokens
SKILL.md length
322 words
Files
1
Skills in repo
15
Repo updated
First seen
Licence
MIT

At a glance

Logs AI thoughts and decisions for human observability. An agent skill from YougLin-dev/Aha-Loop.

  • Works in 4 steps: Log significant thoughts and decisions… → Provide visibility into AI reasoning… → Create anchors for humans to understand… → …
  • Tasks that involve Observability
  • SKILL.md covers Workspace Mode Note, The Job, When to Log and Log Format, plus 6 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Observability is an agent skill from YougLin-dev/Aha-Loop. Logs AI thoughts and decisions for human observability. Applies continuously throughout all tasks to maintain transparency.

Its SKILL.md is about 1.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in DevOps & Cloud, covering Observability. The repository describes itself as: [MVP] Aha Loop is a fully autonomous AI development system, extended from the core ideas of Ralph. It's not just an execution engine, but a complete AI development framework with… The licence is MIT.

When your agent uses it

  • Tasks that involve Observability

Example prompts

  • “/observability”

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Log significant thoughts and decisions to logs/ai-thoughts.md
  2. Provide visibility into AI reasoning process
  3. Create anchors for humans to understand what's happening
  4. Never hide failures or uncertainty

What it can do on your machine

Read from SKILL.md and the folder at commit 8d799b2. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Observability loads about 1.7k tokens when it runs. Until then it costs about 34 tokens; SKILL.md has 322 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~34
When it runs · the whole SKILL.md, loaded when a task matches
~1.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from YougLin-dev/Aha-Loop at commit 8d799b2, republished under its MIT licence (© YougLin-dev). 322 words, ~1,672 tokens.

Download SKILL.mdSave it as .claude/skills/observability/SKILL.md (or your agent's skills folder).
name
observability
description
Logs AI thoughts and decisions for human observability. Applies continuously throughout all tasks to maintain transparency.

Observability Skill

Maintain transparent, human-readable logs of AI decision-making process.

Workspace Mode Note

When running in workspace mode, log to .aha-loop/logs/ai-thoughts.md instead of logs/ai-thoughts.md. The orchestrator will provide the actual paths in the prompt context.


The Job

  1. Log significant thoughts and decisions to logs/ai-thoughts.md
  2. Provide visibility into AI reasoning process
  3. Create anchors for humans to understand what's happening
  4. Never hide failures or uncertainty

When to Log

Always Log
  • Starting a new task - What am I about to do and why
  • Major decisions - When choosing between approaches
  • Unexpected findings - Something surprising or important
  • Errors and recovery - What went wrong and how I'm handling it
  • Completion - What was accomplished
Optional Log (Detailed Level)
  • Intermediate steps
  • Research findings
  • Minor decisions
  • Progress updates

Log Format

Append to logs/ai-thoughts.md:

markdown
## 2026-01-29 14:30:00 | Task: PRD-003 | Phase: Research

### Context
I'm researching authentication strategies for the web app. This is critical 
because it affects security architecture and user experience.

### Inner Thoughts
The vision mentions "simple" and "quick to use". Traditional username/password 
might add friction. Considering passwordless options but need to evaluate 
complexity tradeoffs.

### Decision Point
- Considering: Traditional email/password
- Considering: Magic link (passwordless)
- Considering: OAuth only (Google/GitHub)
- **Chosen:** Magic link with OAuth fallback
- **Reason:** Aligns with "simple" goal, reduces password fatigue, 
  OAuth provides familiar alternative

### Current Progress
- [x] Read vision requirements
- [x] Research auth options
- [ ] Prototype magic link flow
- [ ] Evaluate email service options

### Observations
- Magic link requires reliable email delivery
- Need to consider rate limiting to prevent abuse
- Session duration is important for UX

### Next Action
Will research email service providers (Resend, SendGrid) and evaluate 
their free tiers since budget constraint mentioned "no paid APIs".

---

Log Levels

Configure in config.json:

Minimal
  • Task start/end only
  • Major errors
Normal (Default)
  • All major decisions
  • Phase transitions
  • Errors and recovery
Detailed
  • Everything above
  • Intermediate thoughts
  • Research notes
  • Minor decisions

Thought Categories

Inner Monologue

Express your actual reasoning:

  • "I'm uncertain about X because..."
  • "This seems risky because..."
  • "I'm choosing Y over Z because..."
  • "I notice that..."
Decision Points

When facing choices:

markdown
### Decision Point
- Considering: [Option A] - [pros/cons]
- Considering: [Option B] - [pros/cons]
- **Chosen:** [Option]
- **Reason:** [Why]
- **Tradeoffs:** [What we're giving up]
Uncertainty

Be honest about what you don't know:

markdown
### Uncertainty
I'm not 100% sure about [X]. My current assumption is [Y] because [Z].
If this proves wrong, I'll need to [fallback plan].
Errors

When things go wrong:

markdown
### Error Encountered
**What happened:** [Description]
**Why:** [Root cause if known]
**Impact:** [What this affects]
**Recovery:** [How I'm handling it]

Principles

Be Transparent
  • Don't hide uncertainty
  • Don't pretend to know what you don't
  • Admit mistakes openly
Be Useful
  • Logs should help humans understand
  • Avoid jargon without explanation
  • Provide context
Be Honest
  • Express genuine reasoning, not performance
  • Include doubts and concerns
  • Note when guessing vs. knowing
Be Concise
  • Important details only
  • No filler text
  • Structured for scanning

Integration

At Task Start
markdown
## [timestamp] | Task: [id] | Phase: Starting

### Context
[What I'm about to do]

### Approach
[How I plan to tackle it]

### Potential Concerns
[What might go wrong]
At Decision Points
markdown
### Decision Point
[Structured decision format]
At Task End
markdown
## [timestamp] | Task: [id] | Phase: Complete

### Summary
[What was accomplished]

### Key Decisions Made
- [Decision 1]
- [Decision 2]

### Learnings
[What I learned that might be useful later]

### Concerns for Future
[Any issues to watch for]

Example: Full Task Log

markdown
## 2026-01-29 10:00:00 | Task: PRD-001 | Phase: Starting

### Context
Beginning the project scaffolding PRD. This sets up the foundation for 
the entire project, so I need to be careful about initial decisions.

### Approach
1. First, review the architecture document for tech stack
2. Initialize the project with chosen tools
3. Set up basic structure
4. Configure development environment

### Potential Concerns
- Version compatibility between dependencies
- Getting the directory structure right from the start

---

## 2026-01-29 10:15:00 | Task: PRD-001 | Phase: Research

### Inner Thoughts
The architecture specifies Rust + Axum. I should verify I'm using the 
latest stable versions. Let me check crates.io...

### Observations
- Axum 0.7.4 is latest (released 2 weeks ago, stable)
- Tokio 1.35 is latest
- Should use workspace structure for future modularity

---

## 2026-01-29 10:30:00 | Task: PRD-001 | Phase: Implementation

### Decision Point
- Considering: Single crate structure
- Considering: Workspace with multiple crates
- **Chosen:** Workspace structure
- **Reason:** Project will likely grow, easier to refactor now than later

### Progress
- [x] Created Cargo.toml with workspace
- [x] Added main crate
- [ ] Add configuration
- [ ] Add basic error handling

---

## 2026-01-29 11:00:00 | Task: PRD-001 | Phase: Complete

### Summary
Successfully set up Rust project with Axum 0.7.4, configured workspace,
added basic structure with src/main.rs entry point.

### Key Decisions Made
- Workspace structure for scalability
- Tokio multi-threaded runtime
- Tracing for logging

### Learnings
- Axum 0.7 changed some APIs from 0.6, needed to use new Router syntax

### Files Created
- Cargo.toml
- src/main.rs
- src/lib.rs

---

Mindset

You are thinking out loud for human benefit.

Your logs should read like a competent colleague explaining their work:

  • Clear reasoning
  • Honest about uncertainty
  • Proactive about potential issues
  • Documenting for future reference

Remember: Humans will read these logs to understand what you did and why. Make their job easy.

© YougLin-dev, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .agents/skills/observability of YougLin-dev/Aha-Loop.

Open the folder on GitHubat commit 8d799b2

Compare with similar skills

Observability next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Observability compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Observability this skillYougLin-dev/Aha-Loop181—~1.7kAutomated safety check: PassMIT
Vercel Optimize Auditvercel-labs/agent-skills32k8 repos~4.3kAutomated safety check: PassNone
Kubeshark Installerkubeshark/kubeshark12k—~3.6kAutomated safety check: NotesApache-2.0
Kubeshark KFL2 Filter Referencekubeshark/kubeshark12k—~3.6kAutomated safety check: PassApache-2.0
KubeSphere ServiceMesh Managerkubesphere/kubesphere17k—~2.4kAutomated safety check: PassCustom licence
Kubernetes Network Root Cause Analysiskubeshark/kubeshark12k—~5.3kAutomated safety check: PassApache-2.0

Similar skills

  • Vercel Optimize Audit

    vercel-labs/agent-skills

    Official

    Runs a metrics-first audit of a deployed Vercel project, gating investigations on real signals to produce ranked, citation-backed cost and performance recommendations.

    32k GitHub starsUsed in 8 repos~4.3k tokens
    DevOps & CloudAuto-check passed
  • Kubeshark Installer

    kubeshark/kubeshark

    Installs and configures Kubeshark on a Kubernetes cluster, choosing between the quick CLI path and a Helm install with custom values.

    12k GitHub stars~3.6k tokensUpdated 2 days ago
    DevOps & CloudAuto-check: notes
  • Syntax reference for KFL2, the CEL-based display filter language used to search Kubernetes network traffic captured by Kubeshark, loaded before any filter is written.

    12k GitHub stars~3.6k tokensUpdated 2 days ago
    DevOps & CloudAuto-check passed
  • KubeSphere ServiceMesh Manager

    kubesphere/kubesphere

    Installs, checks and troubleshoots the KubeSphere ServiceMesh extension (Istio, Kiali, Jaeger), including grayscale release, sidecar injection, topology and tracing issues.

    17k GitHub stars~2.4k tokensUpdated 2 mo ago
    DevOps & CloudAuto-check passed
  • Investigates past Kubernetes incidents from Kubeshark traffic snapshots: takes captures, dissects API calls, extracts PCAPs and compares traffic over time.

    12k GitHub stars~5.3k tokensUpdated 2 days ago
    DevOps & CloudAuto-check passed
  • Caveman Gateway Setup

    JuliusBrussee/caveman

    Routes every LLM call in a repository through the Caveman Cloud gateway in record mode, so requests and costs are measured without changing behavior.

    111k GitHub starsUsed in 1 repo~2.6k tokens
    DevOps & CloudAuto-check: warnings

More from YougLin-dev/Aha-Loop

All 15 skills in this repo
  • Architect

    YougLin-dev/Aha-Loop

    Designs system architecture and selects technology stack based on vision analysis.

    181 GitHub stars~2.2k tokensUpdated 8 mo ago
    Auto-check passed
  • Doc Review

    YougLin-dev/Aha-Loop

    Reviews and cleans up outdated documentation. An agent skill from YougLin-dev/Aha-Loop.

    181 GitHub stars~1.5k tokensUpdated 8 mo ago
    Auto-check passed
  • God Consensus

    YougLin-dev/Aha-Loop

    Guides God Committee members through consensus-building for collective decisions.

    181 GitHub stars~2k tokensUpdated 8 mo ago
    Auto-check passed
  • God Intervention

    YougLin-dev/Aha-Loop

    Guides God Committee members through executing interventions.

    181 GitHub stars~2.3k tokensUpdated 8 mo ago
    Auto-check passed
  • God Member

    YougLin-dev/Aha-Loop

    Defines God Committee member behavior and responsibilities with oversight authority.

    181 GitHub stars~1.6k tokensUpdated 8 mo ago
    Auto-check passed
  • Parallel Explore

    YougLin-dev/Aha-Loop

    Guides parallel exploration of multiple implementation approaches using git worktrees.

    181 GitHub stars~1.9k tokensUpdated 8 mo ago
    Auto-check passed

Categories

Questions about Observability

What does Observability do?

Logs AI thoughts and decisions for human observability. An agent skill from YougLin-dev/Aha-Loop. Observability is an agent skill from YougLin-dev/Aha-Loop. Logs AI thoughts and decisions for human observability.

When should I use Observability?

Observability fits situations like: tasks that involve Observability.

How do I install Observability in Claude Code?

Run `npx skills add YougLin-dev/Aha-Loop --skill observability -a claude-code`. Or copy the skill folder (.agents/skills/observability in YougLin-dev/Aha-Loop) into .claude/skills/observability in your project. Claude Code loads it when a task matches its description.

How do I install Observability in Codex?

Run `npx skills add YougLin-dev/Aha-Loop --skill observability -a codex`. Or copy the skill folder (.agents/skills/observability in YougLin-dev/Aha-Loop) into .agents/skills/observability in your project. Codex loads it when a task matches its description.

Can I use Observability in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add YougLin-dev/Aha-Loop --skill observability -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/observability, .gemini/skills/observability, .github/skills/observability and .opencode/skills/observability in your project.

What does Observability need to run?

SKILL.md names no scripts, command-line tools or credentials: Observability is instructions for the agent only.

Does Observability access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Observability safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Observability use?

Observability is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Observability use?

About 1.7k tokens (SKILL.md is roughly 6.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Observability?

Skills that share tags, products or a category with Observability: Vercel Optimize Audit (vercel-labs/agent-skills, 32k stars), Kubeshark Installer (kubeshark/kubeshark, 12k stars), Kubeshark KFL2 Filter Reference (kubeshark/kubeshark, 12k stars) and KubeSphere ServiceMesh Manager (kubesphere/kubesphere, 17k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Observability?

YougLin-dev (a GitHub user) maintains it in YougLin-dev/Aha-Loop, which has 181 GitHub stars. The repository holds 15 skills in this directory. The repository was last updated on February 3, 2026.

Source: YougLin-dev/Aha-Loop on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.